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.
CEREQ (Centre d'études et de recherche sur les qualifications)
1.2. Contact organisation unit
DFC (Département formation et certification)
1.3. Contact name
Restricted from publication
1.4. Contact person function
Restricted from publication
1.5. Contact mail address
10, Place de la Joliette 13567 Marseille cedex 02
1.6. Contact email address
Restricted from publication
1.7. Contact phone number
Restricted from publication
1.8. Contact fax number
Restricted from publication
21 March 2023
2.1. Metadata last certified
21 March 2023
2.2. Metadata last posted
21 March 2023
2.3. Metadata last update
21 March 2023
3.1. Data description
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).
Visa enquête obligatoire - https://www.legifrance.gouv.fr/jorf/id/JORFTEXT000043861776
6.2. Institutional Mandate - data sharing
Not applicable.
The French component of CVTS is a survey of French public statistics. It is recognized as being of general interest and of statistical quality and law n° 51-711 of June 7, 1951 on the obligation, coordination and secrecy in statistics applies.
7.1. Confidentiality - policy
Computer law and freedom voted in 1978.
Céreq respects the laws and regulations in force. In accordance with the GDPR, Céreq has appointed a DPO in charge of managing the governance of personal data within the establishment. It carries out a mission of information, advice and internal control. He reports directly to the management of the establishment.
Physical access at the establishment level is limited, the server rooms are secure and accessible only to authorized persons.
On the telephone service provider side, it was selected following a call for tenders for which proof of the security of the data processed in terms of GDPR was requested.
7.2. Confidentiality - data treatment
According to the law mentioned in 7.1.
The data collected will also be available to researchers as part of an exploitation group coordinated by Dares, France Competences and Céreq.
The data will be anonymized according to the usual anonymization rules: Internally and for the operating group, all the variables of the operating database allowing the company to be directly identified will be deleted (SIREN number, company name) and collection management (contact details of the company or interlocutors within the company).
For the operating group then for the provision of data in addition to the deletion of identifying variables, the variables identified as being quasi-identifying or being attributes will be subject to aggregations.
Finally, availability of the database incorporating the SIREN number will be communicated to the CASD in order to be able to carry out anonymous and secure pairings with other sources. A request to the secrecy committee is necessary.
The results are published in the collections of the institutions, but also online in dataviz on the websites of Céreq and Dares (the Ministry of Labour).
Responding companies also receive a link to the first results, and the most important stakeholders are invited to a presentation.
8.1. Release calendar
The statistical results of CVTS will be the subject of publications. The first publication for the parliamentary budget debate was released by the Ministry of Labour in early October 2022. The results are then published in the collections of Céreq, the Ministry of Labour (Dares) and France Competences in early 2023, with a press conference and a presentation of the main results to stakeholders in the spring of 2023. Indicators are published with dataviz on the websites of Céreq and the Ministry of Labour (Dares).
8.2. Release calendar access
Not applicable.
8.3. Release policy - user access
Responding companies receive a link to the first results by email. The most important indicators are available online in dataviz on the websites of Céreq and Dares. The results with comments and analyses are published in the collections of the institutions, with a press release. Finally an event is co-organized by Céreq, Dares and France Compétences for a presentation of the results to the most important stakeholders and their representatives.
Every 5 years.
The main indicators and metadata are published on the websites of Céreq and the Ministry of Labour (Dares).
10.1. Dissemination format - News release
Press releases are issued when the publications come out and a press conference is planned for the spring 2023 on the occasion of an event co-organized by Céreq, Dares and France Compétences for a presentation of the results to the most important stakeholders and their representatives.
10.2. Dissemination format - Publications
The publications in 4 pages of the institutions: Céreq Bref, Dares Analyses, France competence Note d'étude, Insee bilan formation emploi.
10.3. Dissemination format - online database
The results with data visualization are available on the Céreq (https://www.cereq.fr/) and Dares (https://dares.travail-emploi.gouv.fr/) websites.
10.3.1. Data tables - consultations
Not applicable.
10.4. Dissemination format - microdata access
Micro-data are accessible for the 3 institution partners for EFE-e (French CVTS) that means Céreq, Dares (Ministry of Labour) and France Compétences (National authority for financing and regulating vocational training and apprenticeship).
Micro-data will be made available to researchers within a exploitation of data group coordinated by Céreq. In a third time, at the end of 2023, micro-data will be accessible after submission at the Committee on Statistical Confidentiality with a secure access via a platform.
The data will probably be put on a “secure bubbles” infrastructure like the CASD. With such a platform, a secure access computer box allows remote access to a secure infrastructure where confidential data is sanctuarized.
This place of data storage and processing is called a “secure bubble”. Coupled with the “waterproof” central IT infrastructure, it forms a coherent set of services controlled from start to finish, guaranteeing the very high level of security for data producers.
10.5. Dissemination format - other
None.
10.5.1. Metadata - consultations
Not applicable.
10.6. Documentation on methodology
To date there are no such documents yet. Later they will be available on the Céreq site www.cereq.fr.
10.6.1. Metadata completeness - rate
Not applicable.
10.7. Quality management - documentation
Later they will be available on the Céreq site www.cereq.fr.
The survey benefited from a dedicated website containing all of the survey's presentation documents. Before answering the online questionnaire, companies were invited to read the preparatory documents which specified the nature of the variables requested and their definitions. Frequently asked questions were also online and companies could ask for any clarifications by email or telephone.
11.1. Quality assurance
The CEREQ subcontracted the interrogation to a research company specializing in telephone and internet surveys. CEREQ specifically recruited for this survey a person in charge of monitoring the quality of the survey. It allowed us to follow up daily with instructions to the investigators.
Once the online responses were recorded, they were checked using an automatic program to validate that the mandatory responses were filled in and to check the consistency of the main quantitative variables. When errors were found, a follow-up email was sent to companies encouraging them to correct their answers in an online form. When cleaning the data, the European checking rules were applied, reconciliations were also made with administrative sources from the Ministry of Labour, in particular on variables A4 and A5.
11.2. Quality management - assessment
For a representativeness of the data according to dimensions specific to France (OPCO: skill operators), the number of companies in the sample is much greater than in previous editions. The construction of the sample was done according to a variable of interest (the rate of access to courses was considered more precise than the rate of training companies) and a stratification different from the sampling plan recommended by Eurostat but with the objective of respecting at least the volumes which would have been obtained according to the Eurostat sampling plan. This results in obtaining precision equal to or greater than that recommended by the Eurostat drawing formula. It should be noted that the application of the stratification as recommended in the manual results in a sample which over-represents industry at the expense of services with regard to the structure of the French economy.
12.1. Relevance - User Needs
In France, enterprises are the main contributor to national spending on continuing vocational training, mainly through the training they provide to their employees. Monitoring this expenditure is therefore essential for all institutional partners.
In the absence of administrative data, CVTS makes it possible to capture the training practices of private sector employers and the effects of changes in French vocational training regulations (latest law of 2018). CVTS is included in EFE which includes an annual French component on key quantitative data to track enterprises spending and training practices. The implementation of this system responds to a recurring request from the social partners and the need for annual publication in a parliamentary budget document. It also responds to the need to measure business satisfaction with the OPCO service offer (legal mission of an institutional partner, France Skills). The annual survey is based on the five-yearly European survey EFE-e (French section of CVTS) with an obligation to respond and a panelised dimension which should make it possible to improve the quality of the information collected.
12.2. Relevance - User Satisfaction
The interest of the survey is discussed in a commission of the CNIS (National Council of Statistics) where institutions and stakeholders are represented, which issues an opinion of contribution then during a commission of the label committee which assigns the character mandatory in the survey. The French questionnaire (including European questions) had also been the subject of upstream consultations with the social partners, representatives of HR associations and accountants, as well as OPCO (skills operators).
12.3. Completeness
No missing sector.
12.3.1. Data completeness - rate
Not applicable.
13.1. Accuracy - overall
The sample was drawn in a stratified and random manner from a sampling frame with no known lack of coverage. In addition, the response rate was obtained following a multi-mode collection (internet and telephone) reducing bias linked to the mode of collection. Finally, the weighting used a non-response model which over-represents companies with a low probability of response.
13.2. Sampling error
The drawing of the probability type sample was carried out according to a stratification of the companies by size and OPCO to which they belong. Within these strata, the objective was to obtain a precision of plus or minus 4 points on the rate of access to CVT courses based on the previous edition (CVTS 5). An over-stratification then made it possible to respect during this draw the structure by size and sector of the sampling frame.
The coefficients of variation were calculated according to a procedure taking into account the sampling design.
13.2.1. Sampling error - indicators
See table 13.2.1 "Sampling errors - indicators" in annex "FR - QR tables CVTS 2020 (excel)".
13.3. Non-sampling error
See items 13.3.1 - 13.3.5.
13.3.1. Coverage error
The survey base is managed by Insee (the national statistics institute) and considered to be exhaustive without duplicates. However, the size of the companies in this register does not always correspond to the size declared by the companies for CVTS. This is how 142 companies registered with less than 10 employees declared having crossed this threshold and entered the scope of CVTS.
See table 13.3.1 "Coverage error" in annex "FR - QR tables CVTS 2020 (excel)".
13.3.1.1. Over-coverage - rate
See table 13.3.1.1 "Over-coverage - rate" in annex "FR - QR tables CVTS 2020 (excel)".
13.3.1.2. Common units - proportion
Not applicable.
13.3.2. Measurement error
Given the exceptional circumstances during the first test due to the health crisis, the questionnaire and the survey protocol were tested twice. This test made it possible to validate the proper functioning of the platform and the award filters.
It should be noted that a large part of the questionnaire has not been modified, the good understanding of European qualitative questions has already been validated during CVTS 5.
The questionnaire was developed on a CAWI platform. The contact details of the enterprises taken from the survey base were enriched by various sources (administrative, webscrapping) in order to target those responsible for continuing education. The enterprises received a first letter and an email for companies where it was filled in with an identifier to connect to the platform and an incentive to prepare their responses. For this preparation, a web page dedicated to the survey was created. Enterprises could find there: a presentation of the survey, its context, its objectives; help in preparing responses with detailed description of quantitative questions and definitions; a general FAQ on the survey; contacts by telephone or email; the link to the online questionnaire. Enterprises could also obtain information by telephone using a number dedicated to the survey. Enterprises that did not respond were contacted by telephone and email. For the dedicated telephone number, for telephone follow-ups and telephone surveys, the interviewers have received specific survey training.
A continuous monitoring program of the quantitative data recorded, checked the data and made it possible to follow up by e-mail with certain enterprises which had not responded correctly online, either with inconsistent answers or non-responses, by giving them a link to a new specific online questionnaire. Finally, given the mandatory nature of the survey, the largest non-responding enterprises received reminders with incentives to respond in order to avoid paying a fine.
13.3.3. Non response error
Only the total non-response was taken into account and modelled in the calculation of the weights.
13.3.3.1. Unit non-response - rate
See table 13.3.3.1 "Unit non-response - rate" in annex "FR - QR tables CVTS 2020 (excel)".
13.3.3.2. Item non-response - rate
See table 13.3.3.2 "Item non-response - rate" in annex "FR - QR tables CVTS 2020 (excel)".
13.3.4. Processing error
Collection was automated CAWI and CATI, coding was also automated, thus limiting errors.
13.3.5. Model assumption error
Not applicable.
14.1. Timeliness
Average data production time: from February 2020 to June 2022.
It should be noted that between April and September 2020 there was a period of uncertainty and waiting for the official response from Eurostat after the countries' request for postponing CVTS in view of the health crisis.
14.1.1. Time lag - first result
21 months for the first results in a parliamentary budget document.
14.1.2. Time lag - final result
Not applicable.
14.2. Punctuality
Data collection was delayed from the original schedule. Indeed, it remained suspended between April and September 2020 following Eurostat's response confirming that the year 2020 was maintained as the reference year and that the postponement requested by the countries was not granted.
See table 14.2 "Project phases - dates" in annex "FR - QR tables CVTS 2020 (excel)".
14.2.1. Punctuality - delivery and publication
Not applicable.
15.1. Comparability - geographical
Concepts and definitions for CVTS 6 are set in the legislation and in the implementation manual, no divergence of national concepts and definitions from the European requirements.
Some additional variables/information related to COVID-19 were collected, see annexes.
See table 15.1 "Comparability - geographical" in annex "FR - QR tables CVTS 2020 (excel)".
15.1.1. Asymmetry for mirror flow statistics - coefficient
Not applicable.
15.2. Comparability - over time
Requirements for CVTS 6 remained unchanged compared to CVTS 5. No changes in the French CVTS 6 implementation that have an impact on the comparability with CVTS 5.
See table 15.2 "Comparability - over time" in annex "FR - QR tables CVTS 2020 (excel)".
15.2.1. Length of comparable time series
Not applicable.
15.3. Coherence - cross domain
See table 15.3 "Coherence - cross-domain" in annex "FR - QR tables CVTS 2020 (excel)".
15.3.1. Coherence - sub annual and annual statistics
Not applicable.
15.3.2. Coherence - National Accounts
Not applicable.
15.4. Coherence - internal
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.
Overall costs excluding labour costs approximately 357 000 euros excluding tax, for all data including French extension to small businesses and health, education and agriculture sectors.
The time taken to complete the questionnaire is difficult to estimate: the size of the questionnaire varies greatly according to the profile of the training/non-training enterprises. The average processing time of the interrogations by CATI is 39 minutes. The connection time for CAWI on the site indicated by the respondents (57 minutes on average) does not count the phases of research and preparation over the questions.
It is a long survey because it includes many quantitative variables that are not always recorded in this format by French enterprises. In order to reduce the questioning time, enterprises are first contacted to find the right contact and encourage them to prepare the necessary documents. Depending on whether the right person has indeed been appointed to respond and the degree of preparation, the investigation time varies greatly.
17.1. Data revision - policy
Not applicable.
17.2. Data revision - practice
Not applicable.
17.2.1. Data revision - average size
Not applicable.
18.1. Source data
The data set is based on a sample survey.
See table 18.1 "Source data and data collection" in annex "FR - QR tables CVTS 2020 (excel)".
18.2. Frequency of data collection
Every 5 years.
From this edition, some questions will be filled in each year for the French component.
18.3. Data collection
The data collection was multi-mode, first self-administered via the internet (CAWI), then for non-respondents via a telephone follow-up which mainly resulted in the company responding to the survey via the internet. Very few companies responded to the survey by telephone. The questionnaire was tested with 400 companies using the same protocol.
See also table 18.1 "Source data and data collection" in annex "FR - QR tables CVTS 2020 (excel)".
18.4. Data validation
The consistency of the main labour cost and training expenditure indicators by stratum was validated with administrative sources.
18.5. Data compilation
Some key variables were imputed in the event of non-response or aberrant data. The method used is that of the hot deck: a donor company was randomly selected from companies of the same size and same sector of activity. Then the ratio variable to be imputed / another variable of this donor company (the 2 variables having a proven link) was assigned to the recipient company in order to calculate the variable to be imputed according to its response to the other variable. This method makes it possible both to maintain variability in the imputation carried out and also to take into account the responses of the recipient company to four responses which are correlated.
Concerning the calculation of weights, a method in 3 stages was carried out: use of the inverse of the probability of drawing (weight of drawing in the sampling base), probability of answering via a non-response model integrating variables linked to the collection (contact details available) and finally a final setting on the structure of the survey base (size, sector, OPCO of membership).
18.5.1. Imputation - rate
See table 18.5.1 "Imputation - rate" in annex "FR - QR tables CVTS 2020 (excel)".
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).
Enterprise definition is compliant with Council Regulation (EEC) No 696/93.
The target population covers 214 896 enterprises.
CVTS 6 covers enterprises with 10 or more persons employed belonging to certain NACE categories. No deviation.
NB: In EFE-e, the French component of CVTS, enterprises between 1 and 9 persons have been added.
Variable A2tot (persons employed) refers to 31 December 2020.
Metropolitan France. DROM-COM (overseas departments and regions and overseas collectivities) are not included.
The reference year for CVTS 6 is the calendar year 2020.
The sample was drawn in a stratified and random manner from a sampling frame with no known lack of coverage. In addition, the response rate was obtained following a multi-mode collection (internet and telephone) reducing bias linked to the mode of collection. Finally, the weighting used a non-response model which over-represents companies with a low probability of response.
Number, EUR.
Some key variables were imputed in the event of non-response or aberrant data. The method used is that of the hot deck: a donor company was randomly selected from companies of the same size and same sector of activity. Then the ratio variable to be imputed / another variable of this donor company (the 2 variables having a proven link) was assigned to the recipient company in order to calculate the variable to be imputed according to its response to the other variable. This method makes it possible both to maintain variability in the imputation carried out and also to take into account the responses of the recipient company to four responses which are correlated.
Concerning the calculation of weights, a method in 3 stages was carried out: use of the inverse of the probability of drawing (weight of drawing in the sampling base), probability of answering via a non-response model integrating variables linked to the collection (contact details available) and finally a final setting on the structure of the survey base (size, sector, OPCO of membership).
The data set is based on a sample survey.
See table 18.1 "Source data and data collection" in annex "FR - QR tables CVTS 2020 (excel)".
Every 5 years.
Average data production time: from February 2020 to June 2022.
It should be noted that between April and September 2020 there was a period of uncertainty and waiting for the official response from Eurostat after the countries' request for postponing CVTS in view of the health crisis.
Concepts and definitions for CVTS 6 are set in the legislation and in the implementation manual, no divergence of national concepts and definitions from the European requirements.
Some additional variables/information related to COVID-19 were collected, see annexes.
See table 15.1 "Comparability - geographical" in annex "FR - QR tables CVTS 2020 (excel)".
Requirements for CVTS 6 remained unchanged compared to CVTS 5. No changes in the French CVTS 6 implementation that have an impact on the comparability with CVTS 5.
See table 15.2 "Comparability - over time" in annex "FR - QR tables CVTS 2020 (excel)".
