Statistical Office of the Republic of Serbia

Department of Education, Science and Culture

Milana Rakića 5

11000 Belgrade

The Continuing Vocational Training Survey (CVTS) collects information on enterprises’ investment in the continuing vocational training of their staff. Continuing vocational training (CVT) refers to education or training measures or activities which are financed in total or at least partly by the enterprise (directly or indirectly). Part financing could include the use of work-time for the training activity as well as financing of training equipment.

Information available from the CVTS is grouped around the following topics:

- Provision of CVT courses and other forms of CVT (training/non-training enterprises)

- CVT strategies

- Participants in CVT courses

- Costs of CVT courses

- Time spent in CVT courses

- Characteristics of CVT courses

- Assessment of CVT activities

The CVTS also collects some information on initial vocational training (IVT).

For further information see the CVTS 6 legislation (http://ec.europa.eu/eurostat/web/education-and-training/legislation) and the CVTS 6 implementation manual (http://ec.europa.eu/eurostat/web/education-and-training/methodology).

The main groupings for enterprises are by economic activity (NACE), size group and training/non-training enterprises.

B to N and R to S of NACE Rev. 2, no deviation of the national survey.

Definitions as well as the list of variables covered are available in the CVTS 6 implementation manual (http://ec.europa.eu/eurostat/web/education-and-training/methodology).

CVTS variables show the existence, amount and costs of any vocational training and courses by an enterprise for their persons employed.

Enterprise. The definition is fully compliant with Council Regulation (EEC) No 696/93.

CVTS 6 covers enterprises with 10 or more persons employed belonging to certain NACE categories (see 3.3). The total number of enterprises of the target population is 15 225.

The number of persons employed (A2tot) refers to the 31^st December.

The entire territory of the Republic of Serbia, excluding Kosovo and Metohija (UN governed territory under UNSCR 1244/1999).

Year 2020.

Not applicable.

Number, euros.

The reference year for CVTS 6 is the calendar year 2020.

At European level

Basic legal act: Regulation (EC) No 1552/2005 of the European Parliament and the Council

Implementing act: Commission Regulation (EU) No 1153/2014, amending Commission Regulation (EC) No 198/2006

At national level:

CVTS 6 is part of the adopted “The Official Statistics Plan for 2021”.

Not applicable.

Articles 44-49 of Official Statistical Law (Official Gazzete of RS, number 104/09) prescribes the legislative measures on confidentiality policy.

Annexes:
Official Statistical Law, Confidentiality; Articles 44-49

See 7.1.

The responsible producers of official statistics keep an up-to-date and publicly available ‘release calendar’. This ‘release calendar’ should be available to the public before the beginning of the calendar year, for the entire following year. The ‘release calendar’ provides an overview of all statistical releases and publications with a precise title, reference period and date of publication.

The official release of the main CVTS results is published at SORS website in the first quarter of 2022.

Official statistics are published in SORS’ website at certain times during the working day, in order to ensure consistency and impartiality in the publication of statistics.

https://www.stat.gov.rs/publikacije/

The SORS provides advice and instructions to users on the availability of results in the entire statistical system, sets up and maintains databases that are the result of official statistics, prepares statistical releases and publications that contain data from all areas of statistics.

The website, including the database, is the main channel for disseminating statistics, based on the results of official statistics. All statistical products (results of statistical surveys accessed by end users) are prepared in electronic form and published on the site free of charge. Printed editions are available to users in the SORS’ library. The results of official statistics are available to all users on equal terms at the same time, and any privileged access by external users, prior to their publication, is not possible. Electronic data storage should be in formats suitable for easy access and not be dependent on special applications. Access to individual data without identifiers (anonymized microdata) can be provided in accordance with legal regulations and with the application of certain procedures. Information on methods and classifications used in the production of official statistics (metadata) and information on the quality of statistical products and services should be available and accessible to the general public and in accordance with the criteria of European statistics.

Every 5 years.

Press releases in the first quarter of 2022.

No publications yet.

Online database at http://ec.europa.eu/eurostat/data/database and at SORS website.

Not applicable.

Eurostat publishes both aggregated data on their website and provides access to anonymised micro-data (scientific use files).

Anonymised micro-data on national level will be provided on users' request.

Not applicable.

Not applicable.

Entire methodology was based on Eurostat's guidelines and recommendations which also can be found at Eurostat's website.

Not applicable.

This quality report.

Following the example of the European Statistical System, and in accordance with the mission and vision, as well as the Quality Policy, the Statistical Office of the Republic of Serbia strives for full harmonization of its statistical production with the European Statistics Code of Practice (CoP).

The European Statistics Code of Practice was adopted by the European Statistical System Committee in 2005. Statistical authorities, including the Commission (Eurostat), national statistical institutes and other national institutions responsible for the development, production and dissemination of European Statistics, together with governments, ministries and the European Council, have committed themselves to complying with this Code.

The quality management system of the Statistical Office of the Republic of Serbia is fully adapted to statistical needs because it systematically improves the quality of statistical processes, final results and data, as well as providing services to users.

For further information please see http://ec.europa.eu/eurostat/web/quality/quality-reporting, in particular the ESS Handbook for Quality reports.

The survey was carried out according to the methodological guidelines and the accompanying instruments provided by Eurostat (CVTS standard questionnaire, CVTS manual, classifications, codebook, checking rules etc.). This was the first CVT survey conducted in the Republic of Serbia. Overall data quality can be considered as satisfactory, having in mind the conditions which marked the whole survey process (high impact of COVID-19 pandemic and uncertain situation in which many enterprises found themselves). The biggest issue were question about total number of hours worked in the reference year by persons employed and total paid working time spent on all CVT courses. However, for any doubt the respondents had been contacted for clarification or confirmation.

It is expected that various users will be interested in CVTS data (some inquiries about final data were already received). Being the first CVTS in the Republic of Serbia, it is still not possible to fully conclude to which specific users (interested parties) and to which extent the data obtained from the survey will be used.

No such measure yet.

Data collected for CVTS 6 cover all variables requested by the CVTS legislation, for the target population as specified in the legislation (see 3.6).

Not applicable.

Given the circumstances (ongoing COVID-19 pandemic), CVTS 6 data can be considered as quite reliable. All stages of the survey were fully respected with all included (methodologists, interviewers, controllers, IT colleagues) putting a maximum effort to obtain quality data. No major discrepancies had been noted. All issues that had emerged during the process was tackled in order to gain the best out of the survey.

A stratified simple random sample was selected. Stratification was according to NACE Rev. 2 categories defined in the EU Regulation, according to size and territory (NUTS 2). Number of persons employed is auxiliary variable used for defining size classes. Exhaustive units are units with 250 and more persons employed and take-all units determined applying Hidiroglou algorithm. The sample was allocated using the Bethel algorithm (optimal allocation). The number of persons employed was used as an auxiliary variable. The planned error is 2% for the estimate of the number of employees on the level of categories NACE Rev. 2 and 2.5% for the categories crossed with the size class. Gross sample size is 6749.

See table 13.2.1 "Sampling errors - indicators" in annex "RS - QR tables CVTS 2020 (excel)".

See 13.3.1 to 13.3.5. below.

Sampling frame was defined using the Statistical Business Register (SBR, state February 2021). The sampling frame contains all enterprises with the main activity in the following sections (based on EU Regulation):

B - Mining and quarrying (05, 06, 07, 08, 09)

C - Manufacturing industry (10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32, 33)

D - Power supply (35)

E - Water supply (36, 37, 38, 39)

F - Construction (41, 42, 43)

G - Wholesale, retail trade and sale of motor vehicles (45, 46, 47)

H - Traffic and storage (49, 50, 51, 52, 53)

I - Accommodation and catering services (55, 56)

J - Information and communication (58, 59, 60, 61, 62, 63)

K - Financial activities (64, 65, 66)

L + M + N + R + S - (68,69,70,71,72,73,74,75,77,78,79,80,81,82,90,91,92,93,94,95,96)

The sampling frame includes units covering listed sections, which submitted a Financial Report for the year 2019. Inactive units and entrepreneurs are excluded.

According to EU Regulation, units with 10 or more employees were included in the sampling frame. Data on number of employees are used from the Central Register of Compulsory Social Insurance (average number of employees in last 12 months). After analysing the coverage of the number of employees by categories listed in EU Regulation (20 NACE Rev. 2 categories) it was concluded that frame covers only 49.7% of employees in the category K66 - Ancillary activities in the provision of financial services and insurance. For this reason, all units from this activity were included in the frame.

No additional cut-off or corrections were made.

The sample frame contains 15225 units. Stratified simple random sample is selected. Gross sample size is 6749 units.

See table 13.3.1 "Coverage error" in annex "RS - QR tables CVTS 2020 (excel)".

See table 13.3.1.1 "Over-coverage - rate" in annex "RS - QR tables CVTS 2020 (excel)". 

Not applicable.

The most frequent errors and issues were:

A2. What was the total number of persons employed by the enterprise? Though it was suggested that average number of persons employed during the reference year would be a better measure, SORS stack to the number of persons employed on 31st December, as this turned out to be easier for enterprises to provide/confirm an accurate data, and it is also in line with the Regulation.

A4. In 2020, what was the total number of hours worked by persons employed for the enterprise? Enterprises had problem to give a reliable number of hours due to frequent work from home.

C3. In 2020, what was the total paid working time (in hours) spent on ALL CVT courses broken down by internal or external CVT courses? Also caused some doubt as numerous courses had been organized online and it was difficult to keep track for each participant and his/her amount of time spent in training, especially in bigger enterprises.

C7. In 2020, what were the costs incurred by the enterprise for the provision of CVT courses? Costs for provision of courses had often been left out in responses, so they had to be sent back for supplement.

Electronic questionnaire was applied by implementing full logical and numerical control that prevented a majority of errors. Data entry took into account rules for skipping questions depending on the respondent's answer.

After the defined period given to enterprises to fill the questionnaire, the first reminder was sent electronically with another appeal to respond. They were given a choice to fill it either electronically or via telephone/postal if it suits them better that way. Those who still did not respond were sent two more urgencies that the survey is mandatory. Besides, enterprises were contacted by telephone with same reminder.

The response rate of the survey was 70.4%. The non-response rate was 29.6% of which vast majority was non-contact – 87.4%, then refusal – 8.9%, while inability to respond and rejected questionnaires contained 3.7% of non-response.

See table 13.3.3.1 "Unit non-response - rate" in annex "RS - QR tables CVTS 2020 (excel)".

See table 13.3.3.2 "Item non-response - rate" in annex "RS - QR tables CVTS 2020 (excel)".

All questions and variables were coded following the instructions from CVTS manual. There are no significant points to be emphasized.

Not applicable.

The reference year for CVTS is 2020. The data were available by the end of 2021.

Results on national level were published at the end of 2021.

Results on national level were published at the end of 2021.

See table 14.2 "Project phases - dates" in annex "RS - QR tables CVTS 2020 (excel)".

The full dataset had been delivered respecting the deadline given in the implementation monitor (October 2021).

Not applicable.

See table 15.1 "Comparability - geographical" in annex "RS - QR tables CVTS 2020 (excel)".

No additional variables related to COVID-19 were collected.

Not applicable.

See table 15.2 "Comparability - over time" in annex "RS - QR tables CVTS 2020 (excel)".

Since this is the first CVTS conducted in Republic of Serbia, there is no previous waves of the survey to be compared with.

Not applicable.

See table 15.3 "Coherence - cross-domain" in annex "RS - QR tables CVTS 2020 (excel)".

Not applicable.

Not applicable.

CVTS results for a given reference year are based on the same microdata and results are calculated using the same estimation methods, therefore the data are internally coherent.

Highly consistent.

FTE for staff involved in the survey - 2300 hours.

Costs of fieldwork - 7200 Euros.

Average time for answering the questionnaire - between 20 and 25 minutes.

Not applicable.

Not applicable.

Not applicable.

See table 18.1 "Source data and data collection" in annex "RS - QR tables CVTS 2020 (excel)".

Data are based on sample survey.

The sample frame contains 15225 units. A stratified simple random sample is selected. The gross sample size is 6749 units.

Every 5 years.

Prior to data collection, announcement letters containing a brief description and goal of the survey were sent to the selected enterprises (via email or post office). The survey was mandatory, based on Articles 26 and 52 of Law on Official Statistics (“Official Gazette RS”, num. 104/09). The selected interviewers and controllers passed a necessary training with detailed insight to the content of questionnaire, data entering application and better understanding of certain concepts. After the pilot testing, its analysis and impressions, the data collection took place between March and June 2021, using a combination of methods (CAWI, CATI and postal). The collected data were revised and coded in accordance with the methodological requirements of the Eurostat and in accordance with valid classifications.

The main tool for data collection was the electronic questionnaire, for which permission needed to be granted to reporting units by sending request for it. It was also used by interviewers doing the telephone interview (CATI). However, these web questionnaires were not the same because web questionnaires had been exclusively for external use (for enterprises who had sent a request to fill it online), while data collected via e-mail, telephone or postal were entered through the internal system application. In both ways, data were checked for logical or numerical inconsistencies. Coding and complete control of the collected data are performed after the data entry process by the supervisor directly in the central database located on the server of the Statistical Office of the Republic of Serbia.

See also table 18.1 "Source data and data collection" in annex "RS - QR tables CVTS 2020 (excel)".

The web questionnaire was created and operated by IT experts of SORS, with all necessary numerical and logical checks. The selected enterprises who opted for this survey method had to send a request to open an account so they could access the web questionnaire and fill it. SORS’s IT experts also created an application on the local Statistical Office’s network in which data that had been collected using e-mail and telephone/postal method were entered, respecting all logical or numerical rules. In both ways, data have ended up in the unique database.

After the defined period given to enterprises to fill the questionnaire, the first reminder was sent electronically with another appeal to respond. They were given a choice to fill it either electronically or via telephone/postal if it suits them better that way. Those who still did not respond were sent two more urgencies that the survey is mandatory. Besides, enterprises were contacted by telephone with same reminder. Thorough control of the received questionnaires was carried out and respondents were contacted in order to clarify or correct any potential incorrect or inconsistent answer.

The final dataset was checked against the checking rules as defined in the CVTS manual.

Design weights are adjusted for non-response. Estimation is standard for stratified simple random sampling. Estimates are provided for variables as well as for their standard errors. For calculating estimates and standard errors we used Swedish ETOS based on software SAS. Estimation is carried out with the following characteristics:

• Strata in which no units in the net sample and the planned sample size is more than one are compressed with non-census strata of the same activity class and the same or neighbouring class by number of employees.
• Detection and treatment of outliers is performed. Units with significant number of employees (250+) are separated from the census stratum to which they belong if the stratum has units that did not responded. They are assigned to the initial weight, 1. For the strata, from which they are excluded, the number of units and number of sample units are calculated again.

No imputations were performed.

See table 18.5.1 "Imputation - rate" in annex "RS - QR tables CVTS 2020 (excel)".

Not applicable.

Not applicable.

The main problem in organizing of survey was an on-going COVID-19 pandemic. Republic of Serbia is one of the countries highly affected by the virus and it made a significant impact on all spheres of life. Many companies and entrepreneurs, especially smaller ones, had been forced to either reduce their work or totally shut it down. Also, staff involved in all aspects of the survey made the highest effort to be at disposal during the entire process.

Having this in mind, the overall organization and obtained results could be considered as satisfactory and the best at the given moment.