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).
Kind of Data
Sample survey data [ssd]
All persons 15-74 years old with permanent residence in Estonia.
Producers and sponsors
Authoring entity/Primary investigators
Statistical Office of Estonia
Statistical Office of Estonia
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.
In 2003, the response rate was 80.9%
The weights are formed in a sequence of steps. A weight resulting from the previous step is multiplied by the correction factor calculated at the current step. The correction factors are scaled in such a way that their sample average is unity at each step. As a result, the final weight is a product of the initial weight and correction factors.
As stratified sampling is used as the first step of sample formation, first the initial weight that is inversely proportional to the inclusion probability in each stratum is calculated.
For non-response adjustment the non-response correction factors are computed. The weighting groups of reasonably uniform size of sampled households are formed on the basis of the place of residence of the household according to the non-response rate in the region. Within each group the correction factor is inversely proportional to the overall response rate in the region.
In the next step the weights are calibrated so that they produce exact population numbers in certain subgroups known from demographic data (including institutional population). For working-age persons the subgroups by sex, age (5-years age groups) and the place of residence (urban/rural area, 15 counties (NUTS-4 level) and the capital city) are considered. For this purpose the linear consistent weighting method is applied.
For nonworking age persons the non-calibrated household’s weights are calibrated by sex and 5-years age groups.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
Type of 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).
Statistical Office of Estonia
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.
The use of the datasets must be acknowledged using a citation which would include:
- the identification of the Primary Investigator (including country name)
- the full title of the survey and its acronym (when available), and the year(s) of implementation
- the survey reference number
- the source and date of download (for datasets disseminated online).
Statistical Office of Estonia. Estonia Labour Force Survey (LFS) 2003, Ref. EST_2003_LFS_v01_M. Dataset downloaded from [URL] on [date].
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.