Labour Force Survey 2003 - Estonia

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

The Estonian Labour Force Survey started in 1997 as an annual survey. Since 2000 the Estonian LFS has been organised as a continuous quarterly survey. The survey covers the whole country. Both private and collective households are surveyed.

Estonia Labour Force Survey provides population estimates for the main labour market characteristics, such as employment, unemployment, inactivity, hours of work, occupation, economic activity and much else, as well as important socio-demographic characteristics, such as sex, age, education, household characteristics and regions of residence. Estonia LFS uses methodology of the International Labour Organisation (ILO), which guarantees the comparability of the data.

The sample size per quarter is approximately 2,300 households. The sampling frame is based on the 2000 Population and Housing Census database, comprising all registered persons 15-74 years old.

Since 2001, the Estonian Labour Force Survey questionnaire includes an ad hoc module, the contents of which vary from year to year. The module is compiled in accordance with the relevant EU regulations. The aim of the added module is to gather detailed information about the aspect of life directly relevant to the labour market, which is comparable in all EU Member States. In 2003, the ad hoc module was lifelong learning (2nd and 4th quarter).

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

National

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

All persons 15-74 years old with permanent residence in Estonia.

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

Sample survey data [ssd]

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

The sample size per quarter is approximately 2,300 households.

Since the year 2000 the sample of the Labour Force Survey was handled in two parts:
- the address-sample where all the working-age members of the household (households) living on the address selected into the sample were interviewed;
- the person-sample where all the working-age members of the sample person's household were interviewed.

In 2002 the transition to the whole person-sample began. All households participating in the survey for the first time were handled by the person-rule. Since the 2nd quarter of 2003 the whole sample is person-sample.

The sample design is stratified systematic sampling of individuals whose households are included into sample. Different inclusion probabilities of households are taken into account during calculation of design weights. Stratification is made by place of residence. The 15 counties of Estonia and Tallinn are divided into four strata according to the population size (I - Tallinn, II - four bigger counties (Harju (without Tallinn), Ida-Viru, Pärnu, Tartu), III - ten smaller counties (Jõgeva, Järva, Lääne, Lääne-Viru, Põlva, Rapla, Saare, Valga, Viljandi, Võru), IV - Hiiu county) and different inclusion probabilities are used in strata, the highest being for Hiiu county.

The sampling frame is based on the 2000 Population and Housing Census database, comprising all registered persons 15-74 years old.

In the first phase the sample is selected by systematic sampling inside strata and the information on the size of sampled households is collected. In the second phase the sample is grouped by the number of persons aged 15-74 years in the household of the sampled individual. The final sample is then selected by systematic sampling from each size group with inclusion probability inverse to the size. This yields an equal probability sample of households (and its 15-74 years old members) inside strata. All persons aged 15-74 years in the households of the final sample are interviewed. Every sampled household is interviewed for four quarters according to the rotation pattern 2-(2)-2.

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

Face-to-face [f2f]

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

The main objective of the Estonian Labour Force Survey is to get data about participation of population of Estonia in the labour market; therefore the main part of the questionnaire includes questions concerning the respondents’ activities in the reference week. The employed persons are asked about economic activity of the main job, occupation, labour status, work relations, place of work, usual and actual working time, about secondary jobs, etc. The unemployed persons are asked about the steps taken to find a job, the duration of job seeking, the characteristics of previous job and a job they are looking for, etc. The persons who do not work and do not look for a job (the inactive persons) are asked about the reason for inactivity, sources of subsistence, etc.

Eurostat’s ad hoc module included in the Estonian Labour Force Survey 2003 was Lifelong learning (2nd and 4th quarter, Section J).

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

Starting with 2000 until the full implementation of CAPI-interviewing in the 4th quarter of 2005 the entry of the data was performed centrally. The co-ordinators of the interviewers' network collected the filled questionnaires from interviewers after every two weeks and sent the paper questionnaires to Statistics Estonia. In case of inconsistencies that appeared during data entering, the interviewer who had made a mistake was contacted and the data were adjusted. After the end of the field work and entering of all data, the Blaise data files were converted into SAS format and more complicated relations, which could not be made in Blaise, were checked. In case of inconsistencies corrections were made again by contacting the interviewer.

Since 2001, the data entry program was improved so that in case of inconsistencies operators had a possibility to add a short comment about the nature of the error to the database as a link to the corresponding question of the questionnaire. The aim of it was to analyze the errors made by interviewers and find out the reasons for that (lack of the interviewer's knowledge, unclear questions, etc.). The co-ordinators of the interviewers' network received the quarterly report of mistakes and dealt with the interviewers whose work quality was low.

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

In 2003, the response rate was 80.9%

摘要
---------------------------
爱沙尼亚劳动力调查始于1997年，最初作为一项年度调查。自2000年起，爱沙尼亚劳动力调查转变为连续的季度调查。该调查覆盖全国范围，对私人及集体家庭进行调查。
爱沙尼亚劳动力调查提供了关于主要劳动力市场特征的人口估计，例如就业、失业、非活动状态、工作时间、职业、经济活动等，以及重要的社会经济特征，如性别、年龄、教育、家庭特征和居住地区等。爱沙尼亚劳动力调查采用国际劳工组织（ILO）的方法论，保证了数据的可比性。
每个季度的样本量约为2,300个家庭。抽样框架基于2000年人口和住房普查数据库，包括所有注册的15至74岁的人口。
自2001年以来，爱沙尼亚劳动力调查问卷包括一个临时模块，其内容每年都有所不同。该模块根据相关欧盟法规编制。附加模块的目的是收集与所有欧盟成员国劳动市场直接相关的生命方面的详细信息。
2003年，临时模块为终身学习（第二季度和第四季度）。
地理覆盖范围
---------------------------
全国
总体
---------------------------
所有在爱沙尼亚拥有永久居留权的15至74岁的人。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
每个季度的样本量约为2,300个家庭。
自2000年起，劳动力调查的样本分为两部分处理：
- 地址样本，其中所有居住在选定的地址样本中的工作年龄家庭成员（家庭）都接受了访谈；
- 个人样本，其中所有样本个人家庭的成年家庭成员都接受了访谈。
2002年开始了向全员样本的转变。所有首次参与调查的家庭都按照个人规则处理。自2003年第二季度起，所有样本都是个人样本。
样本设计是对包括在样本中的家庭成员的分层系统抽样。在计算设计权重时考虑了家庭的不同纳入概率。分层是根据居住地进行的。爱沙尼亚的15个县和塔林市根据人口规模分为四个层次（I - 塔林，II - 四个大县（哈留（不含塔林）、伊达-维鲁、帕尔努、塔尔图），III - 十个小县（约格瓦、耶尔瓦、拉恩、拉恩-维鲁、波尔瓦、拉普拉、萨雷、瓦尔加、维连迪、沃鲁），IV - 赫尤县）并在层次中使用不同的纳入概率，赫尤县的概率最高。
抽样框架基于2000年人口和住房普查数据库，包括所有注册的15至74岁的人口。
在第一阶段，样本在层次内通过系统抽样选择，并收集关于样本家庭规模的信息。在第二阶段，样本根据样本个人家庭中15至74岁的人数进行分组。然后，从每个规模组中通过系统抽样选择样本，纳入概率与规模成反比。这产生了一个在层次内家庭（及其15至74岁成员）的等概率样本。最终样本家庭中的所有15至74岁的人都接受了访谈。每个样本家庭根据轮换模式2-(2)-2接受四个季度的访谈。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
爱沙尼亚劳动力调查的主要目标是获取关于爱沙尼亚人口在劳动力市场中的参与情况的数据；因此，问卷的主要部分包括关于受访者参考周活动的问题。受雇者被问及主要工作的经济活动、职业、劳动状态、工作关系、工作地点、通常和实际工作时间、关于第二职业等。失业者被问及寻找工作的步骤、求职的持续时间、先前工作的特征以及他们正在寻找的工作的特征等。不工作也不找工作的人（非活动人员）被问及非活动的原因、生计来源等。
包含在2003年爱沙尼亚劳动力调查中的欧盟临时模块为终身学习（第二季度和第四季度，第J节）。
数据清理操作
---------------------------
从2000年开始，直到2005年第四季度完全实施CAPI访谈，数据输入是在中心进行的。访谈员网络协调员在每两周后从访谈员那里收集填写的问卷，并将纸质问卷发送到爱沙尼亚统计局。在数据输入过程中出现不一致时，会联系犯错访谈员，并调整数据。在实地工作和所有数据输入结束后，Blaise数据文件被转换为SAS格式，并检查了Blaise中无法实现的关系。在出现不一致的情况下，会再次通过联系访谈员进行更正。
自2001年以来，数据输入程序得到了改进，以便在出现不一致时，操作员有权限将关于错误性质的简短评论添加到数据库中，作为与问卷相应问题的链接。其目的是分析访谈员所犯的错误，并找出错误的原因（访谈员知识缺乏、问题不明确等）。访谈员网络协调员收到了关于错误的季度报告，并处理了工作质量低下的访谈员。
响应率
---------------------------
2003年的响应率为80.9%