Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
The quarterly data are broken down by economic activity (at section level) in accordance with NACE Rev. 2 - Statistical classification of economic activities in the European Community.
2.3. Coverage - sector
The data cover all the economic activities defined by NACE Rev. 2, except agriculture, forestry and fishing activities (optional in the JVS framework regulation), and the activities of households as employers and the activities of extraterritorial organisations and bodies.
2.4. Statistical concepts and definitions
A 'job vacancy' is defined as a paid post that is newly created, unoccupied, or about to become vacant:
for which the employer is taking active steps and is prepared to take further steps to find a suitable candidate from outside the enterprise concerned; and
which the employer intends to fill either immediately or within a specific period of time.
‘Active steps to find a suitable candidate’ include:
notifying the job vacancy to the public employment services,
contacting a private employment agency/head hunters,
advertising the vacancy in the media (for example internet, newspapers, magazines),
advertising the vacancy on a public notice board,
approaching, interviewing or selecting possible candidates/potential recruits directly,
approaching employees and/or personal contacts,
using internships.
'Specific period of time’ refers to the maximum time the vacancy is open and intended to be filled. That period shall be unlimited; all vacancies for which active steps are continuing on the reference date shall be reported.
An 'occupied post’ means a paid post within the organisation to which an employee has been assigned.
2.5. Statistical unit
The statistical unit for the data collection is the enterprise for NACE sections B to S except groups 85.1, 85.2 and 85.3 (in NACE section P) for which the statistical unit is the local unit.
2.6. Statistical population
Enterprises with one and more employees.
2.7. Reference area
Belgium and NUTS 1 regions (Brussels-Capital Region, Flemish Region and Walloon Region).
2.8. Coverage - Time
Data are from the first quarter 2010 onwards.
2.9. Base period
Not applicable.
3.1. Source data
Identification of the source of the data
The statistical processing method uses two data sources: the monthly Dimona flows (administrative source from the NSSO, National Social Security Office) and the quarterly data from the enterprise web survey. The monthly Dimona flows are only used for sampling and extrapolation of survey results.
Coverage
Geographical
The whole territory of the country
NACE
The data cover all the economic activities defined by NACE Rev. 2, except agriculture, forestry and fishing activities (optional in the JVS framework regulation), and the activities of households as employers and the activities of extraterritorial organisations and bodies.
Enterprise size
The whole population of enterprises with 1+ employees.
Definition of the statistical unit
The statistical unit for the data collection is the enterprise for NACE sections B to S except groups 85.1, 85.2 and 85.3 (in NACE section P) for which the statistical unit is the local unit.
Remarks
Sampling design
Base used for the sample
DBRIS enterprise database
Sampling design
The number of units sampled is determined by the stratum population size with minimum sample size thresholds depending on the employment size class. This implies exhaustive sampling for large employers.
Retention/renewal of sampling units
There is a rotating panel approach. Depending on the inclusion probability, rotation is annual (max 3 consecutive years) or quarterly (4 quarters).
Sample size
11,017 enterprises (on average): 10,890 in the 1st quarter, 10,995 in the 2nd quarter, 11,053 in the 3rd quarter and 11,130 in the 4th quarter.
674 local units (on average): 673 in the 1st quarter, 669 in the 2nd quarter, 677 in the 3rd quarter and 677 in the 4th quarter.
Stratification
Strata are composed of employer groups of eligible NACE 1 (section level) crossed with employment size class (1-9).
Other sources
Maintenance agency
Dimona (employment) database
Updating frequency
Monthly
Rules for clearance (of outdated information)
Fresh data for month include consecutive updates for months n-2 and n-1.
Voluntary/compulsory reporting and sanctions
Voluntary reporting
Remarks
In 2020 (starting in mid-march), the Covid-19 crisis severely impacted the Belgian labour market. In order to reduce the Covid-19 spread, the Belgian government decided to close a large number of companies (mainly in I, R and S sections) during several limited periods of time. To reduce the workload on businesses, Statbel did not send any survey to those temporarily closed companies (for example restaurants in 2021Q1). Therefore, the values of sections that were impacted by a temporary closure were supplemented by administrative data coming from regional public employment services (Forem for Wallonia, VDAB for Flanders and Actiris for Brussels). Anyway, since the third survey of 2021, as there was no more temporary closed companies, the usual survey has been used to estimate the whole sections.
3.2. Frequency of data collection
Reference dates
Dimona flows are monthly administrative data from the NSSO (National Social Security Office). These data are transmitted on the third Thursday of each month for the three previous months. The data for the first two months are updated data
The web survey is carried out every quarter with the middle of the quarter as reference date: 15 February 2024 (first quarter), 15 May 2024 (second quarter), 15 August 2024 (third quarter) and 15 November 2024 (fourth quarter).
3.3. Data collection
Brief description of the data collection method(s)
Remarks
For each transmission, Dimona flows from the NSSO (National Social Security Office) are stored in a single table in which the latest monthly data are saved.
The data collection process is carried out in several phases. Letters are first sent between two and five days before the reference date. Survey forms are then completed by enterprises (or local units) between the reference date and the next two and a half months. Two reminder letters are issued in order to increase the response rate.
The reform launched in 2016 and completed in 2019 led to two achievements for the first quarter of 2020: 1) a simplification of the questionnaire for enterprises with occupied posts in only one region, 2) an additional module with two additional questions about the duration of job vacancies and the recruitment channels.
This additional module is only attached to the first survey (Q1) of each year since 2020.
The statistical processing method uses the monthly Dimona flows (administrative source from the NSSO, National Social Security Office) for the pre-filling, the sampling, the extrapolation and the calibration. It uses the quarterly data from the enterprise (or local unit) web survey to estimate the number of occupied posts and job vacancies by sector and enterprise size.
3.4. Data validation
Data validation is performed at two levels.
Individual data are validated by the Data Collection Department with the help of the Economic Statistics Department. Aggregated data are then validated by the Economic Statistics Department and the Planning Department.
3.5. Data compilation
Brief description of the weighting method
Weighting dimensions
In order to reflect the number of employees and job vacancies in enterprises, weights are calculated on the basis of the total number of jobs within enterprises in the stratum and among respondents from the sample rather than on the basis of their number. Through this method we overcome a potential bias that might be due to an unrepresentative sample or a selective non-response. In this case, the average size of employment of respondents in the stratum may not be representative of the stratum average. The risk of bias is inversely proportional to the number of employers drawn in each stratum.
See point 3.1, sampling design
3.6. Adjustment
The seasonally adjusted series on the number of occupied posts and job vacancies have been produced for all NACE sections B to S using the processing model RSA4 of the JDemetra+ software.
4.1. Quality assurance
Not applicable.
4.2. Quality management - assessment
Since 2012 the Belgian Job Vacancy Survey has been used to provide the estimated number of formal and informal job vacancies (published by regional employment offices or not) on a quarterly basis, by enterprise size (more or less than 10 employees) and economic sector (NACE sections from B to S). From 2013 onwards, the survey also allows estimations of occupied posts and job vacancies for temporary positions.
Since 2014 the number of occupied posts and job vacancies in sections O, P and Q has been estimated in three phases: i) determination of results extrapolated from the survey among enterprises, ii) breakdown of data from phase i) by section (O, P and Q) on the basis of allocation keys for occupied posts, iii) integration of the results extrapolated from the survey among local units into section P (only for NACE groups 85.1, 85.2 and 85.3).
Phase ii) is carried out in order to take into account the fact that many local units belonged to enterprises that were not classified in the same NACE section. This was, for example, the case for municipal schools (P) belonging to municipal administrations (O) or for university hospitals (Q) belonging to universities (P). Phase iii) was required when most enterprises classified in NACE groups 85.1, 85.2 and 85.3 (compulsory and subsidised education) indicated that they did not know the number of job vacancies in their local units.
The seasonally adjusted series on the number of occupied posts and job vacancies have been produced for all NACE sections B to S using the processing model RSA4 of the JDemetra+ software.
5.1. Relevance - User Needs
Description of the national users and their main needs
Remarks
Federal institutions:
National Bank of Belgium (NBB)
Federal Public Service (FPS) Employment, Labour and Social Dialogue
Central Economic Council (CEC)
Regional institutions:
Regional employment offices (Actiris, Le Forem, VDAB, ADG)
Institut de recherches économiques et sociales (IRES)
The estimated number of job vacancies at national and regional level
The institutions listed above are the members of the working party that fostered the development of the Belgian Job Vacancy Survey.
Other institutions show an interest in JVS data, but do not take part in this working party.
The Job Vacancy Survey is intended to meet the needs of Eurostat and national users (regional employment offices, National Bank of Belgium, FPS Employment,...).
5.2. Relevance - User Satisfaction
Extent to which the needs of national users are satisfied (voluntary)
Remarks
Since May 2015 the needs of national users have been completely satisfied as JVS data for the year 2014 are published by enterprise size, by quarter, by NACE section (B to S) and at national and regional level.
All data for 2024 (including regional data) were published at the time of the transmission to Eurostat.
5.3. Completeness
Description of missing variables and missing breakdowns of the variables
Report progress on the implementation measures, including:
a detailed plan and timetable for completing implementation
a summary of the remaining deviations from EU concepts
The absolute number of occupied posts and job vacancies at national and regional level is provided.
Data on occupied posts and job vacancies by province (NUTS-2 level) and according to the classification of occupations (ISCO-08 classification) have not been published yet.
A method for estimating the number of occupied posts and job vacancies by province and according to the classification of occupations has been studied as part of the JVS Grant 2012 (transmission done in March 2014).
The reform launched in 2016 was completed in 2019 and no solution could be found to estimate occupied posts and job vacancies by province and according to the classification of occupations.
5.3.1. Data completeness - rate
90% of the quartely data were delivered.
Only the data according to the classification of occupations have not been delivered yet.
6.1. Accuracy - overall
The Job Vacancy Survey underestimates the number of occupied posts and job vacancies in NACE sections B and R, which may account for the differences with LFS estimates for these sections.
The Job Vacancy Survey underestimates the number of occupied posts and job vacancies in NACE section R as many local units in this section belong to enterprises classified in section O.
The Job Vacancy Survey also underestimates the number of occupied posts and job vacancies in section B due to the size and the geographic concentration (above all in Wallonia) of this sector.
6.2. Sampling error
That was studied by NACE sector.
6.2.1. Sampling error - indicators
Coefficient of variation (taking into account the sampling design) or estimated sampling error for the number of job vacancies (see guidelines).
NACE sector
1st quarter
2nd quarter
3rd quarter
4th quarter
B
0.120625
0.132571
0.112172
0.146791
C
0.009721
0.007675
0.006646
0.006451
D
0.010311
0.014093
0.010574
0.059527
E
0.047616
0.031642
0.032398
0.032961
F
0.015000
0.019984
0.020838
0.015618
G
0.014567
0.015993
0.012424
0.030807
H
0.037379
0.034189
0.032219
0.031020
I
0.030570
0.034521
0.029356
0.028435
J
0.025862
0.021940
0.022845
0.027210
K
0.024430
0.030787
0.027102
0.023995
L
0.040788
0.033653
0.030582
0.044345
M
0.017849
0.015982
0.017409
0.019630
N
0.035680
0.031958
0.020745
0.024047
O
0.036405
0.026184
0.038446
0.036930
P
0.052117*
0.121050*
0.083796*
0.078195*
Q
0.008157
0.007214
0.007490
0.007247
R
0.059499
0.035246
0.021915
0.032760
S
0.022544
0.023085
0.016462
0.029708
Total (B-S)
0.007154
0.006495
0.006328
0.007692
* : only for NACE groups 85.4 and 85.5. The values of other groups was not transmitted because calculated with another survey.
6.3. Non-sampling error
Information on variables with non-negligible measurement and processing errors
Information on main sources of (non-negligible) measurement and processing errors and, if available, on methods applied for correction
Estimation bias: An assessment of the non-sampling errors, in terms of the absolute number of vacant posts, for the total number of job vacancies and, where possible, for aggregation level of NACE Rev. 2 specified in Annex 1 to this Regulation and size classes (1-9, 10 + employees).
Remarks
Underestimation of occupied posts and job vacancies
If we observe the differences in the estimation of occupied posts between JVS and LFS, the underestimation is 11.0% (or 450,000 occupied posts), which represents an underestimation of 20,000 job vacancies. This underestimation would be higher for sectors D (59.0%), P (49.0%) and R (44.0%).
6.3.1. Coverage error
Description of any difference between the reference population and the study population
Description of classification errors
Description of any difference between the reference dates and the reference quarter
Any other relevant information
Discrepancies can be found between the study population and the reference population for occupied posts according to the NSSO (National Social Security Office).
Different definitions between the survey and the NSSO (National Social Security Office).
Reporting errors in the survey and to the NSSO (National Social Security Office).
6.3.1.1. Over-coverage - rate
Over-coverage – rate
No over-coverage
6.3.1.2. Common units - proportion
The occupied posts are also in the administrative databases of the NSSO (National Social Security Office) but the definitions are slightly different. The only great difference between the two sources is that the temporary posts are classified in the temporary employment agencies for administrative data and in the businesses for the survey.
6.3.2. Measurement error
Although enterprises can (in)validate pre-filled data on occupied posts (on the basis of Dimona flows from the NSSO/NSSOPLA), reporting errors may occur in the number of occupied posts and job vacancies. Enterprises may be misled by pre-filled data, especially if these data are close to the real value. Reporting errors may also occur when declarations are not checked as reported data are sufficiently close to the pre-filled value.
Data checks are indeed performed if the relative difference between pre-filled values and survey data is more than 25%. In 30% of cases, these differences are precisely due to reporting errors. For job vacancies, data are only checked when enterprises reported more job vacancies than occupied posts. In these few cases, 50% of declarations contain errors.
6.3.3. Non response error
When enterprises respond to the survey, all questions contain an answer as the survey is relatively simple (only 12 answers to give: the number of occupied posts and job vacancies for permanent and temporary positions, for the three regions).
6.3.3.1. Unit non-response - rate
Unit response rate
Unit response rate = number of respondent enterprises / number of selected enterprises in the sample
Enterprise survey (NACE sections B to S, excluding NACE groups 85.1, 85.2 and 85.3)
First quarter = 6,362/ 10,890 = 58.42%
Second quarter = 6,259/ 10,995 = 56.93%
Third quarter = 6,583/ 11,053 = 59.56%
Fourth quarter = 6,531/ 11,130 = 58.68%
Local unit survey (NACE groups 85.1, 85.2 and 85.3)
First quarter = 287/ 673 = 42.64%
Second quarter = 284/ 669 = 42.45%
Third quarter = 269/ 677 = 39.73%
Fourth quarter = 307/ 677 = 45.35%
6.3.3.2. Item non-response - rate
Respondent enterprises provide an answer for all items.
6.3.4. Processing error
No processing error.
6.3.4.1. Imputation - rate
Item imputation rate and methods and, where possible, the effect of imputation on the estimates for the variables transmitted
In 2020 (starting in mid-march), the Covid-19 crisis severely impacted the Belgian labour market. In order to reduce the Covid-19 spread, the Belgian government decided to close a large number of companies (mainly in I, R and S sections) during several limited periods of time. To reduce the workload on businesses, Statbel did not send any survey to those temporarily closed companies (for example restaurants in 2021Q1). Therefore, the values of sections that were impacted by a temporary closure were supplemented by administrative data coming from regional public employment services (Forem for Wallonia, VDAB for Flanders and Actiris for Brussels). Anyway, since the third survey of 2021, as there was no more temporary closed companies, the usual survey has been used to estimate the whole sections.
6.3.5. Model assumption error
If modelling is used, include a description of the models used. Particular emphasis should be given to models for imputation or grossing-up to correct for unit non-response.
No model assumption error
6.4. Seasonal adjustment
Brief description of seasonal adjustment procedures, in particular with regard to the European Statistical System guidelines on seasonal adjustment which have been endorsed and supported by the SPC.
The seasonally adjusted series on the number of occupied posts and job vacancies have been produced for all NACE sections B to S using the processing model RSA4 of the JDemetra+ software.
Provide a revision history, including the revisions in the published number of job vacancies and a summary of the reasons for the revisions.
No revision
6.6.1. Data revision - average size
Not applicable.
7.1. Timeliness
Data are delivered to Eurostat 70 days after the reference period. The data were published at national level for the first time in October 2014. These data were for 2012 and 2013. The data for 2014 to 2016 were published two weeks after transmission to Eurostat. Data from 2017 to 2019 were published at the time of the transmission to Eurostat. Since 2020, the data are published 1 or 2 days after the transmission to Eurostat.
The annual results at national level are published at the same time as the national publication of the 4th quarter results.
7.1.1. Time lag - first result
Information on the time span between the release of data at national level and the reference period of the data.
The quarterly publications of the 2024 data are published one or two days after the transmission to Eurostat (70 days after the end of the reference period).
7.1.2. Time lag - final result
Same situation as in 7.1.1 because the publication is only once.
7.2. Punctuality
A workflow is implemented to ensure data transmission to Eurostat within 70 days after the end of the reference period.
7.2.1. Punctuality - delivery and publication
Deadlines for the respondents to reply, also covering recalls and follow-ups
Period of the fieldwork
Period of data processing
Dates of publication of first results
Remarks
The survey is launched at the middle of the quarter (e.g.: 15 February 2024 for the first quarter).
Enterprises respond within two and a half months after the survey was launched (e.g.: survey ends on 30 April 2024 for the first quarter.
The workflow is implemented two months before the launch of the survey (e.g.: 15 December 2023 for the first quarter)
During the third month after the launch of the survey (e.g.: in May 2024 for the first quarter of 2024)
Data validation in the first week of the fourth month after the survey was launched (e.g.: the first week of June 2024 for the first quarter).
Data transmission to Eurostat on the 70th day after the end of the period reference (e.g.: 09 June 2024 for the first quarter).
8.1. Comparability - geographical
Information on differences between national and European concepts, and — to the extent possible — their effects on the estimation.
No difference as the survey has been primarily designed to meet the needs of Eurostat. European concepts have been applied.
The data are comparable for occupied posts with publications from the NSSO (National Social Security Office) and the Labour Force Survey (excluding sections B and R). For job vacancies, the data from the regional employment offices (Actiris, ADG, Le Forem and the VDAB) are too different and partial.
8.1.1. Asymmetry for mirror flow statistics - coefficient
The survey only studies the occupied and vacant posts in Belgium.
The survey is not interested in the frontier workers.
8.2. Comparability - over time
Information on changes in definitions, coverage and methods in any two consecutive quarters, and their effects on the estimation.
Remarks
There is a first break between 2011 and 2012 as there is a shift from the use of administrative sources to the production of survey data (see reports of the JVS Grants 2010, 2011 and 2012).
A second break can be found between 2012 and 2013 with the integration of temporary positions (see two reports of the JVS Grants 2012).
A third break can be identified between 2013 and 2014 due to the change in methodology for the estimation of occupied posts and job vacancies in sections O, P and Q.
8.2.1. Length of comparable time series
The years 2012 to 2014 for permanent positions in sections B to S excluding sections O, P and Q.
The years 2013 and 2014 for temporary and permanent positions in sections B to S excluding sections O, P and Q.
From 2014 up to now for temporary and permanent positions in all sections B to S.
8.3. Coherence - cross domain
Comparisons of data on the number of vacant jobs from other relevant sources when available, in total and broken down by NACE at section level when relevant, and reasons if the values differ considerably.
Data on job vacancies published by the regional employment offices (Actiris, ADG, Le Forem and VDAB) are different. All these offices use an estimation method that is specific to their region and is based on the number of job offers officially reported to their agencies. These offices therefore do not know the number of unofficial offers.
The Job Vacancy Survey underestimates the number of occupied posts and job vacancies in NACE sections B and R, which may account for the differences with LFS estimates for these sections. Attached you can find a comparison of occupied posts between data of the JVS and the LFS.
Quarterly and annual data on occupied posts from the NSSO (National Social Security Office), as well as data from the Labour Force Survey, are comparable with the survey data, except sections B and R (see in 6.1 for explanations).
Survey data on job vacancies are in no way comparable with the data published by the regional employment offices on the number of job offers recorded in their agencies.
8.5. Coherence - National Accounts
National Accounts use the quarterly and annual data on occupied posts from the NSSO (National Social Security Office). Job vacancies are not estimated. However, unemployment statistics are mainly based on the data from the regional employment offices.
8.6. Coherence - internal
The results by sector and enterprise size estimated on the basis of the survey are coherent with each other.
9.1. Dissemination format - News release
Not planned any more.
9.2. Dissemination format - Publications
The analysis is disseminated in downloadable Word format from our "Statbel" website.
Dissemination scheme, including to whom the results are sent
Periodicity of national publication
References for publications of core results, including those with commentary in the form of text, graphs, maps, etc.
Information on what results, if any, are sent to reporting units included in the sample
Dissemination of results on our "Statbel" website.
Our main clients receive an e-mail when the results are disseminated.
70 days after the reference period.
An analysis presenting the principal results of the Job Vacancy Survey.
Not applicable
9.3. Dissemination format - online database
Not planned yet.
9.3.1. Data tables - consultations
Dissemination of a single downloadable Excel file on our "Statbel" website in which all annual and quarterly data are contained.
9.4. Dissemination format - microdata access
No dissemination of individual data.
9.5. Dissemination format - other
Not planned yet.
9.6. Documentation on methodology
Publication of a webpage on our "Statbel" website containing a presentation of the main methodological characteristics of the Job Vacancy Survey.
9.7. Quality management - documentation
All documentation on the Job Vacancy Survey is archived on our servers as part of a "Synergy" project (which is a programme to reform the workflow at five levels: standardisation, optimisation, documentation, dissemination and quality).
Description of and references for metadata provided
References for core methodological documents relating to the statistics provided
Description of main actions carried out by the national statistical services to inform users about the data
Remarks
Two Word documents on metadata:
on the collection;
on the statistical production.
All methodological documents are archived as part of a "synergy" project.
Participation in meetings of user working parties (see 5.1).
Presentation at a workshop of the CEC (Central Economic Council).
9.7.1. Metadata completeness - rate
Update of two Word documents on metadata in December of every year: one on the collection and another on the statistical production.
9.7.2. Metadata - consultations
Publication of a summary on our "Statbel" website containing relevant information from the updated documents under 9.7.
Production costs and the response burden are reduced:
The survey form is quite short and simple (only 12 answers to give);
The use of a web survey;
Data on occupied posts are pre-filled on the basis of recent administrative data (Dimona flows from the NSSO, National Social Security Office).
The reform launched in 2016 and completed in 2019 led to two achievements in the first quarter of 2020: 1) a simplification of the questionnaire for enterprises with occupied posts in only one region, 2) an additional module with two additional questions about the duration of job vacancies and the recruitment channels.
This additional module is only attached to the first survey (Q1) of each year since 2020.
11.1. Confidentiality - policy
Individual data are protected by the Belgian Statistics Act. The publication at national level may only include aggregated results.
11.2. Confidentiality - data treatment
Disclosure rules: Brief description of when data have to be deleted for reasons of confidentiality
There are no specific rules applied during the data treatment. The identification numbers of the Crossroads Bank for Employment (common identification numbers for all public authorities) are stored throughout the data treatment.
Confidentiality rules are less stringent for enterprises than for natural persons. However, individual data (per enterprise) need to be anonymised when they are transmitted outside Statistics Belgium.
Special remarks
Seasonally adjusted series on the number of occupied posts and job vacancies were added to the data for 2015 from the third quarter.
A 'job vacancy' is defined as a paid post that is newly created, unoccupied, or about to become vacant:
for which the employer is taking active steps and is prepared to take further steps to find a suitable candidate from outside the enterprise concerned; and
which the employer intends to fill either immediately or within a specific period of time.
‘Active steps to find a suitable candidate’ include:
notifying the job vacancy to the public employment services,
contacting a private employment agency/head hunters,
advertising the vacancy in the media (for example internet, newspapers, magazines),
advertising the vacancy on a public notice board,
approaching, interviewing or selecting possible candidates/potential recruits directly,
approaching employees and/or personal contacts,
using internships.
'Specific period of time’ refers to the maximum time the vacancy is open and intended to be filled. That period shall be unlimited; all vacancies for which active steps are continuing on the reference date shall be reported.
An 'occupied post’ means a paid post within the organisation to which an employee has been assigned.
The statistical unit for the data collection is the enterprise for NACE sections B to S except groups 85.1, 85.2 and 85.3 (in NACE section P) for which the statistical unit is the local unit.
Enterprises with one and more employees.
Belgium and NUTS 1 regions (Brussels-Capital Region, Flemish Region and Walloon Region).
Not Applicable
The Job Vacancy Survey underestimates the number of occupied posts and job vacancies in NACE sections B and R, which may account for the differences with LFS estimates for these sections.
The Job Vacancy Survey underestimates the number of occupied posts and job vacancies in NACE section R as many local units in this section belong to enterprises classified in section O.
The Job Vacancy Survey also underestimates the number of occupied posts and job vacancies in section B due to the size and the geographic concentration (above all in Wallonia) of this sector.
Not Applicable
Brief description of the weighting method
Weighting dimensions
In order to reflect the number of employees and job vacancies in enterprises, weights are calculated on the basis of the total number of jobs within enterprises in the stratum and among respondents from the sample rather than on the basis of their number. Through this method we overcome a potential bias that might be due to an unrepresentative sample or a selective non-response. In this case, the average size of employment of respondents in the stratum may not be representative of the stratum average. The risk of bias is inversely proportional to the number of employers drawn in each stratum.
See point 3.1, sampling design
Identification of the source of the data
The statistical processing method uses two data sources: the monthly Dimona flows (administrative source from the NSSO, National Social Security Office) and the quarterly data from the enterprise web survey. The monthly Dimona flows are only used for sampling and extrapolation of survey results.
Coverage
Geographical
The whole territory of the country
NACE
The data cover all the economic activities defined by NACE Rev. 2, except agriculture, forestry and fishing activities (optional in the JVS framework regulation), and the activities of households as employers and the activities of extraterritorial organisations and bodies.
Enterprise size
The whole population of enterprises with 1+ employees.
Definition of the statistical unit
The statistical unit for the data collection is the enterprise for NACE sections B to S except groups 85.1, 85.2 and 85.3 (in NACE section P) for which the statistical unit is the local unit.
Remarks
Sampling design
Base used for the sample
DBRIS enterprise database
Sampling design
The number of units sampled is determined by the stratum population size with minimum sample size thresholds depending on the employment size class. This implies exhaustive sampling for large employers.
Retention/renewal of sampling units
There is a rotating panel approach. Depending on the inclusion probability, rotation is annual (max 3 consecutive years) or quarterly (4 quarters).
Sample size
11,017 enterprises (on average): 10,890 in the 1st quarter, 10,995 in the 2nd quarter, 11,053 in the 3rd quarter and 11,130 in the 4th quarter.
674 local units (on average): 673 in the 1st quarter, 669 in the 2nd quarter, 677 in the 3rd quarter and 677 in the 4th quarter.
Stratification
Strata are composed of employer groups of eligible NACE 1 (section level) crossed with employment size class (1-9).
Other sources
Maintenance agency
Dimona (employment) database
Updating frequency
Monthly
Rules for clearance (of outdated information)
Fresh data for month include consecutive updates for months n-2 and n-1.
Voluntary/compulsory reporting and sanctions
Voluntary reporting
Remarks
In 2020 (starting in mid-march), the Covid-19 crisis severely impacted the Belgian labour market. In order to reduce the Covid-19 spread, the Belgian government decided to close a large number of companies (mainly in I, R and S sections) during several limited periods of time. To reduce the workload on businesses, Statbel did not send any survey to those temporarily closed companies (for example restaurants in 2021Q1). Therefore, the values of sections that were impacted by a temporary closure were supplemented by administrative data coming from regional public employment services (Forem for Wallonia, VDAB for Flanders and Actiris for Brussels). Anyway, since the third survey of 2021, as there was no more temporary closed companies, the usual survey has been used to estimate the whole sections.
Not Applicable
Data are delivered to Eurostat 70 days after the reference period. The data were published at national level for the first time in October 2014. These data were for 2012 and 2013. The data for 2014 to 2016 were published two weeks after transmission to Eurostat. Data from 2017 to 2019 were published at the time of the transmission to Eurostat. Since 2020, the data are published 1 or 2 days after the transmission to Eurostat.
The annual results at national level are published at the same time as the national publication of the 4th quarter results.
Information on differences between national and European concepts, and — to the extent possible — their effects on the estimation.
No difference as the survey has been primarily designed to meet the needs of Eurostat. European concepts have been applied.
The data are comparable for occupied posts with publications from the NSSO (National Social Security Office) and the Labour Force Survey (excluding sections B and R). For job vacancies, the data from the regional employment offices (Actiris, ADG, Le Forem and the VDAB) are too different and partial.
Information on changes in definitions, coverage and methods in any two consecutive quarters, and their effects on the estimation.
Remarks
There is a first break between 2011 and 2012 as there is a shift from the use of administrative sources to the production of survey data (see reports of the JVS Grants 2010, 2011 and 2012).
A second break can be found between 2012 and 2013 with the integration of temporary positions (see two reports of the JVS Grants 2012).
A third break can be identified between 2013 and 2014 due to the change in methodology for the estimation of occupied posts and job vacancies in sections O, P and Q.
