The source for the data on the number of occupied posts is monthly survey the Persons in Employment (DAK), and the data for this survey are primary taken from the Statistical Register of Employment (SRDAP). Data refer to the whole population and are not calculated on the basis of a sample. 

The source for persons who are recipients of parental compensation (persons on maternity leave) is the Ministry of Labour, Family, Social Affairs and Equal Opportunities.

The source for persons who are on long-term sick leave more than 30 working days is the National Institute of Public Helth.

Excluded are:

- persons who are recipients of parental compensation (persons on maternity leave) - since January 2008 and

- persons who are on long-term sick leave which is longer than 30 working days - since January 2013.

Data is collected quarterly. Reporting date for job vacancies is the last working day in the midle month of the reference quarter. The reference dates for job vacancies in 2022 were:

The occupied posts refer to the last day in the middle month of the reference quarter. Before 2015 the average of the three months was used (as situation on the last day of the month) for occupied post and for job vacancies as well.

• by the e-STAT (WEB application),
• by CATI (interview by phone),
• by taking over the data collected by the ESS (employers in the public sector and state-owned companies).

Data collection begins on the reference day. In the beginning the reporting units have 14 days to report data via the e-STAT application (WEB). After this first period, a list of units with non-response is prepared. These units we call in the next two weeks to obtain the missing information (CATI implementation).

With weighting adjustment we want to achieve representativeness of the sample, so that the weighted data give us as good population estimates as possible. The process of weighting depends on the sampling design, the unit non-response rate and available auxiliary variables used for calibration. The final weight is the product of the sampling weight, the non-response weight and the calibration factor. The results represents an estimate of the total target population. The process of weighting depends on the sampling design, the unit non-response rate and available auxiliary variables used for calibration. The final weight is the product of the sampling weight, the non-response weight and the calibration factor.

In this survey, the final (population) weight consists of three types of weights: - Sample weight,  - Non-response weight, - Calibration (adjustment to population values).

The first two types of weights are calculated based on the strata from the sample (both size classes and activities) obtained from the frame. The last (population) weight is calculated on the basis of new collected data: statistical activities (SKD_ST_OBD_ID1) and size classes, using the number of occupied posts (ZDM_STEV) from the collected data. The population weight therefore includes quarterly updated information on statistical activity and size classes (strata), calculated on the basis of the number of occupied posts.

Communication with individual reporting units takes place via SURSs Call Centre and with users of statistical data via SURSs Information Centre, directly by telephone, fax or e-mail.

The Employment Service of Slovenia published monthly data as a cumulative sum of the data from the first day in the month to the last day in the month. Included are data for business entities that want to carry out mandatory public disclosure of information through the ESS. Since mid-April 2013, the ESS publishes data on job vacancies reported by employers if they wish to carry out a public notice via their e-services. In this way, collected and published data are not representative and does not allow a comprehensive view of the supply of vacancies in Slovenia. The Statistical Office of Slovenia published quarterly data collected by the sample survey. Because of the mentioned methodological differences, the published data differ from the data published by the ESS. Users have at their disposal the same information, but published by different methodologies depending on what data they need for their further analyses.

SURS measured general user satisfaction for the last time in 2020. Respondents assessed general satisfaction with SURS with the average score of 8.1 (on a scale from 1 – disagree completely to 10 – agree completely). Before that in 2018 it was 7.5.

- a detailed plan and timetable for completing implementation

- a summary of the remaining deviations from EU concepts

There are no deviations from EU concepts.

The time series of the average coefficient of variation by size of the business entities:

Coefficient of variation by fields of activities and by size of the business entities, Slovenia, quartely

Not applicable

The tables show the non-response rate for unweighted and weighted data for the last three survey implementations. The average unweighted non-response rate was the lowest in 2022; it was 3.38% and for weighted data, it was the lowest in 2022; it was 5.09%. The average unweighted non-response rate has been decreasing in the last three years, the highest was in 2020, it was 8.5%. The year 2020 was marked with the covid-19 pandemic, which caused additional difficulties in data collection. Especially in the first quarter of 2020, because we couldn't conduct a telephone survey (CATI); On 12 March 2020, it was declared lockdown in Slovenia with restricted movement of persons and prohibited performing economic activities (or companies were allowed to do so to a limited extent), which lasted until the end of May. The second wave of the lockdown was declared on October 18 and lasted until 2021, but companies have already adapted to the given situation and enabled work from home. In line with the guidelines and standards for communicating with reporting units on minimizing non-response and collecting the highest possible quality of data, SURS reminds the reporting units of their obligation via e-mail by sending them a reminder before the reporting deadline and a reminder. For units that did not respond via eSTAT, SURS tries to obtain data by conducting a CATI survey. SURS has established a call centre which helps the reporting units in case of lack of understanding of the questionnaire, and a list of frequently asked questions with answers is published online. However, there may be some units from which we do not succeed to obtain the data, namely units with invalid addresses and/or invalid telephone numbers. Overall, the non-response rate is low, regarding the large number of units included in this survey, especially micro-enterprises with less than ten employees.

For seasonal adjustment of time series we use the TRAMO/SEATS method. A time series model is set up, and it is revised in detail and corrected about once a year. By using the model, the time series is decomposed into:

- the trend-cycle component (which consists of the trend and cyclical movements over a period longer than one year),

- the seasonal component (which consists of the seasonal effects and the calendar effects; the calendar effects are composed of the working day effect, the leap-year effect, the holiday effect and the Easter effect),

- the irregular component (which consists of random fluctuations and some outliers).

Seasonally adjusted data are original data from which the seasonal and calendar effects are eliminated.

Data on the number of PDM (job vacancies) and ZDM (occupied posts) are seasonally adjusted. The time series according to SKD2008 includes fields of activity from B - S, namely from the first quarter of 2008 onwards.

Models for seasonal adjustment were reviewed in the 3^rd quarter of 2022 and the models were changed for individual areas, some activities were added, and some no longer have a seasonal impact and the models are no longer needed.

Job vacancies seasonal adjustment (PDM)  refer to 2022 Q1 and Q2:

In PDM, 52 time series are seasonally adjusted, 36 time series are seasonally adjusted directly, and 16 time series are indirectly seasonally adjusted (groups at a higher level of activity). 14 directly seasonally adjusted time series are characterized by seasonal and calendar influences (seasonal models are set), and 22 time series are characterized by non-seasonal models, which means that seasonally adjusted data are the same as the original ones.

Job vacancies seasonal adjustment (PDM)  refer to 2022 Q3 and Q4:

In PDM, 52 time series are seasonally adjusted, 36 time series are seasonally adjusted directly, and 16 time series are indirectly seasonally adjusted (groups at a higher level of activity). 12 directly seasonally adjusted time series are characterized by seasonal and calendar influences (seasonal models are set), and 24 time series are characterized by non-seasonal models, which means that seasonally adjusted data are the same as the original ones.

Occupied posts seasonal adjustment (ZDM) refer to 2022 Q1 and Q2:

At ZDM, 52 time series are also seasonally adjusted, 36 time series are seasonally adjusted directly, and 16 time series are indirectly seasonally adjusted (groups at a higher level of activity). 26 directly seasonally adjusted time series are characterized by seasonal and calendar influences (seasonal models are set), and 10 time series are characterized by non-seasonal models, which means that seasonally adjusted data are the same as the original ones.

Occupied posts seasonal adjustment (ZDM) refer to 2022 Q3 and Q4:

At ZDM, 52 time series are also seasonally adjusted, 36 time series are seasonally adjusted directly, and 16 time series are indirectly seasonally adjusted (groups at a higher level of activity). 20 directly seasonally adjusted time series are characterized by seasonal and calendar influences (seasonal models are set), and 16 time series are characterized by non-seasonal models, which means that seasonally adjusted data are the same as the original ones.

Together with the publication of data for the third quarter of 2011, on 24 November 2011, revised data for the entire time series, from 2001 on, were published. In line with the methodology stipulated in the Regulation (EC) No. 453/2008, occupied posts include only persons in paid employment, while before the revision persons in employment (excluding farmers) were included. Since 2009 occupied posts include posted workers (workers sent abroad to work or study) who have employment contracts in Slovenia and whose wages are paid by employers registered in Slovenia.

The next reason for the revision was exclusion of recipients of parental compensation (persons on maternity leave) since 2008. While receiving parental compensation, person on maternity leave formally remain employed and are included in persons in employment. During their leave, employers often employ another person for the same job to replace the absent employee. Excluding persons on maternity leave from the number of occupied posts is thus necessary in order not to duplicate the number of occupied posts.

Data on the number of job vacancies remained more or less the same, except that activities of households as employers are also included.

Taking into account the above mentioned changes, the number of occupied posts was lower; on average by about 50,000 (between 2001 and 31 December 2007) and by about 72,000 (since 2008 when the data on persons on maternity leave are taken into account). As a result, the job vacancy rate also changed; since revision in 2011 it was one percentage point higher.

Revision of the data in 2015:

Together with the release for the first quarter of 2015 on 9 June 2015 revised data for the 2013Q2–2014Q1 period were also published. Previously, those data had been provisional. Because of the elimination of the compulsory reporting of job vacancies, the administrative data source of the ESS became incomplete since the second quarter of 2013. In 2013 SURS was trying to find alternative methods for forecasting these data but without success. In October 2014 SURS conducted a pilot implementation of the survey. Processes and data collection were tested, data collection was carried out by eSTAT application (WEB). In 2015 SURS actually began to collect data on job vacancies with an independent survey.  In the meantime, data were estimated on the basis of historical time series by Demetra+ program with TRAMO/SEATS method.  For the calculation of the forecast data, we used the new available information, i.e. data collected for the first quarter of 2015 (SURS) and results from the survey »Napovednik zaposlovanja« (NAP-ZAP), which was conducted in May and in October 2014 by the ESS. The survey was conducted at employers with 10 or more persons in employment (threshold coverage). SURS had one question at the end of the questionnaire regarding the number of current job vacancies on the reference day, namely: 

-       For the second quarter of 2014 the reference day was 30 April 2014;

-       For the fourth quarter of 2014 the reference day was 31 October 2014.

The sample for the NAP-ZAP survey included nearly 9,000 business entities. The survey rate was 57% in the second quarter of 2014 and slightly lower (53%) in the fourth quarter. The survey included only business entities with 10 or more employees. For the recalculation to the level of Slovenia we used the ratio of 2012 (for the second and fourth quarters of 2012).  Data for other interim quarters were recalculated with Demetra+ program again. Considering the added information, the provisional data greatly changed. 

In 2015 there is a break in the time series due to changes in the data source, the method of data collection and the methodology.

Data is collected at field once per quarter. The reference day for job vacancies is the last working day in the middle month of the quarter. The reference day for occupied posts is the last day in the middle month of the reference quarter.

Data collection ends on the last day of the reference quarter.

Results are published with the First Release, the dates are announced in advance for the whole year in Release Calendar. In 2022, there were no delay by publishing the data.

Dates of publications of first results are listed in the table in chapter 7.1.1. Time lag - first result.

Since 2001 there were no major methodological or legal changes which would have effect to the job vacancies data. The time series job vacancy data is comparable from 2001 to 2014, only the data from the second quarter of 2013 to the end of 2014 were forecasted on the basis of historical time series data and additional information from other surveys. In 2015Q1, SURS began to collect job vacancy data with survey based on representative sample.

The other changes are reflected to occupied post:

Since January 2008 from occupied post are excluded recipients of parental compensation (persons on maternity leave). While receiving parental compensation, person on maternity leave formally remain employed and are included in persons in employment. During their leave, employers often employ another person for the same job to replace the absent employee.

Since September 2009 in the number of occupied posts are includes posted workers (workers sent abroad to work or training) who have employment contracts in Slovenia and whose wages are paid by employers registered in Slovenia. 

Since January 2013 from occupied post are excluded persons on long-term sick leave more than 30 working days. Such sick leave is no longer covered by the employer but by the Health Insurance Institute.

Since 1 January 2020 posts occupied by persons employed by two or more employers at the same time up to the maximum full time (40 hours per week) have been included.

Individual data users, journalist, institutions can sign up to receive daily information about new publications. They also get information when next release is announced.

The data are published quarterly. Since 2013 detail annually data by occupation, by activity and by statistical regions are no longer available. 

Core results are published with the First Release, links are listed in chapter 9.1.

Data is not protected in terms of confidentiality because they are not published at such a detailed level.

In accordance with the principles of confidentiality at SURS, all results are published in aggregate form, as absolute numbers and in percentages. For easier interpretation and understanding of data, we also prepare infographics, graphs, tables, and calculate various indicators.