Research and development (R&D) (rd)

1.1. Contact organisation

Belgian Science Policy Office (Belspo)

1.2. Contact organisation unit

Monitoring and evaluation of research and innovation (MERI)

1.3. Contact name

Restricted from publication

1.4. Contact person function

Restricted from publication

1.5. Contact mail address

Boulevard Simon Bolivar 30, 1000 Brussels, Belgium

1.6. Contact email address

Restricted from publication

1.7. Contact phone number

Restricted from publication

1.8. Contact fax number

Not required.

2.1. Metadata last certified

30 October 2025

2.2. Metadata last posted

31 October 2025

2.3. Metadata last update

31 October 2025

3.1. Data description

Statistics on Business enterprise R&D (BERD) measure research and experimental development (R&D) performed in the business enterprise sector, i.e., R&D expenditure and R&D personnel. In line with this objective, the target population for the national R&D survey of the business enterprise sector consists of all R&D performing enterprises (including all R&D performers – occasional and continuous, known and unknown – in all branches and size classes) belonging to this sector. The “enterprise” is defined in Council Regulation (EEC) No 1993/696 of 15 March 1993. The results are related to the population of all R&D performing enterprises classified in Sections A to U of the common statistical classification of economic activities as established by Regulation (EC) No 1893/2006 of the European Parliament and of the Council (NACE Rev.2).

The main concepts and definitions used for the production of R&D statistics are given by OECD (2015), Frascati Manual, which is the internationally recognised standard methodology for collecting R&D statistics, and by Eurostat’s EBS Methodological Manual.

Since the beginning of 2021, the collection of R&D statistics is based on Commission Implementing Regulation (EU) No 2020/1197 of 30 July 2020. The Regulation sets the framework for the collection of R&D statistics and specifies the main variables of interest and their breakdowns at predefined level of detail.

3.2. Classification system

• The distribution of principal economic activity and by industry orientation are based on Statistical classification of economic activities in the European Community (NACE Rev. 2);
• The local unit for the statistics are compiled at regional level according to NUTS 2 – Nomenclature of Territorial Units for Statistics;
• The distribution by socioeconomic objectives (SEO) are based on Nomenclature for the Analysis and Comparisons of Scientific Programmes and Budgets (NABS);
• The fields of research and development are based on Classification and distribution by Fields of Research and Development (FORD)
• The R&D personnel and researchers by educational attainment are classified by the International Standard Classification of Education ISCED 2011

3.3. Coverage - sector

Please see the sub-concepts 3.3.1 to 3.3.5. in the full metadata view.

3.3.1. General coverage

Definition of R&D

R&D comprise creative and systematic work undertaken in order to increase the stock of knowledge - including knowledge of humankind, culture and society - and to devise new applications of available knowledge.

3.3.2. Sector institutional coverage

Business enterprise sector (BES)	In 1993 (1992 for the business enterprise sector) a new methodology was introduced for compiling the R&D statistical series. This remedied the gaps in the 1980s series (see below). Since then, BERD and BES total R&D personnel numbers comply with Frascati Manual standards. BES includes both public and private enterprises, as well as collective research centres controlled by or primarily serving business enterprises. Over the period 1977-1989, R&D expenditure of collective research centres was underestimated because these centres were not exhaustively surveyed. From 1992 onwards, this was corrected. From 1990 onwards, the R&D effort of public enterprises is rather marginal compared to the total R&D effort of BES, due to privatisations of major public enterprises engaged in R&D. The introduction of the new methodology led to the recalculation of the BERD series and the total personnel series of BES for the period 1992-1995. This re-evaluation of both series led to breaks in the series compared to the data available for 1989 and earlier. No estimates were made for the period 1990-1991. There was a constitutional reform in Belgium in 1990. Because of this reform the production of R&D statistics was moved from the national to the regional level, as the regions gained more authority. Hence, from 1992 onwards, the BES R&D survey is organized by the regions at the NUTS 1 level. In the period 1981-1989, R&D personnel numbers of BES were underestimated, since no information was available for R&D personnel in public enterprises. Since 1992, however, public enterprises are included in the R&D survey of the business enterprise sector, and this survey includes questions on R&D personnel.
Hospitals and clinics	Hospital and medical centres affiliated with universities are included in the higher education sector. Other hospitals do not occur in our sources of potentially R&D active entities, so we do not survey them.
Inclusion of units that primarily do not belong to BES and the borderline cases.	The official Belgian business register is used as a sampling frame. Non-profits and non-market oriented entities are excluded. No divergence from the Frascati Manual guidelines.

3.3.3. R&D variable coverage

R&D administration and other support activities	No divergence from the Frascati Manual guidelines.
External R&D personnel	External R&D personnel were for the first time included in the total R&D personnel numbers for reference year 2021. This caused a break-in-series as in all previous years only internal R&D personnel numbers were reported as total R&D personnel numbers. As required by the EU legislation, since reference year 2021, the total R&D personnel numbers have been the sum of both internal and external R&D personnel.
Clinical trials: compliance with the recommendations in FM §2.61.	Clinical trials phases 1, 2 and 3 are included in the intramural R&D expenditure of the enterprises. Phase 4 clinical trials should be excluded; however, this is difficult to verify.

3.3.4. International R&D transactions

Receipts from rest of the world by sector - availability	Yes, the BES R&D survey covers funding from abroad
Payments to rest of the world by sector - availability	Yes, the BES R&D survey covers funding of extramural R&D abroad
Intramural R&D expenditure in foreign-controlled enterprises – coverage	Yes, the BES R&D survey allows identifying R&D from foreign-controlled affiliates

3.3.5. Extramural R&D expenditures

According to the Frascati Manual, expenditure on extramural R&D (i.e. R&D performed outside the statistical unit enterprise) is not included in intramural R&D performance totals (FM, §4.12).

Data collection on extramural R&D expenditure (Yes/No)	Yes, we collect and compile data on extramural R&D expenditure in the business enterprise sector
Method for separating extramural R&D expenditure from intramural R&D expenditure	We show a chart in the survey form, to help respondents distinguish between the two.
Difficulties to distinguish intramural from extramural R&D expenditure	Difficulties remain. For example, some important Belgian groups have affiliates abroad that conduct R&D under full supervision of the Belgian headquarter, and consider these workers abroad in their survey response as consultants (hired from within the group) for their own in-house R&D; hence the numbers of these affiliates abroad are included in the intramural R&D numbers of the Belgian groups. Other Belgian groups with a less centralized structure consider R&D performed by their affiliates abroad to be extramural R&D projects. In both cases the R&D performed by the workers abroad will also count towards intramural R&D in those home countries (resulting in double counting in the first case: the R&D workers abroad are included in the intramural R&D numbers of the mother firm AND they are included in the intramural R&D numbers of the affiliate abroad).

3.4. Statistical concepts and definitions

Please see the sub-concepts 3.4.1 and 3.4.2 in the full metadata view.

3.4.1. R&D expenditure

Coverage of years	By calendar year
Source of funds	No divergence from Frascati Manual guidelines
Type of R&D	No divergence from Frascati Manual guidelines
Type of costs	No divergence from Frascati Manual guidelines. Since the 2014 survey, we include in our survey a detailed breakdown of capital expenditure in land and buildings, instruments and equipment, computer software, and intellectual property
Economic activity of the unit	No divergence from Frascati Manual guidelines. Since the 2021 reference year, economic activity reflects the economic activity of enterprises as derived from the official business register in Belgium.
Economic activity of industry served (for enterprises in ISIC/NACE 72)	Product field data are collected in our BES R&D surveys
Product field	Product field data are collected in our BES R&D surveys.
Defence R&D - method for obtaining data on R&D expenditure	Given that such expenditure is limited in the case of Belgium, no data are tracked separately for this type of activity. R&D conducted by the Department of Defense is included in the Government R&D numbers, however.

3.4.2. R&D personnel

See below.

3.4.2.1. R&D personnel – Head Counts (HC)

Coverage of years	From reference year 2021 onwards, the total for R&D personnel consists of the sum of internal and external R&D personnel (consultants), as requested by the European legislation. Before 2021 only numbers for internal R&D personnel (own employees) were reported. Hence, in 2021 there is a break-in-series for total R&D personnel.
Function	No divergence.
Qualification	We collect this information in our BES R&D survey, but only for internal R&D personnel, as this information is impossible to obtain for external R&D.
Age	We do not collect age data in our BES R&D survey.
Citizenship	Not collected in our BES R&D survey.

3.4.2.2. R&D personnel – Full Time Equivalent (FTE)

Coverage of years	From reference year 2021 onwards, the total for R&D personnel consists of the sum of internal and external R&D personnel (consultants), as requested by the European legislation. Before 2021 only numbers for internal R&D personnel (own employees) were reported. Hence, in 2021 there is a break-in-series for total R&D personnel.
Function	No divergence.
Qualification	We collect this information in our BES R&D survey, but only for internal R&D personnel, as this information is impossible to obtain for external R&D personnel (consultants).
Age	We do not collect age data in our BES R&D survey.
Citizenship	Not collected in our BES R&D survey.

3.4.2.3. FTE calculation

FTEs are collected directly in the questionnaire.

3.5. Statistical unit

The statistical unit for BERD is the enterprise as defined by Council Regulation (EEC) No 1993/696 of 15 March 1993, if there are deviations please explain.

The statistical unit is the enterprise but the reporting units are legal units.

3.6. Statistical population

Please see the sub-concepts 3.6.1 and 3.6.2 in the full metadata view.

3.6.1. National target population

The objective of the European R&D statistics is to cover all intramural R&D activities. In line with this objective the target population for the national R&D survey of the Business Enterprise Sector should consist of all R&D performing enterprises (including all R&D performers – occasional and continuous, known and unknown - in all branches and size classes) belonging to this sector. In practice however, countries in their R&D surveys might exclude some enterprises for which R&D activities are deemed to be non-existent or negligible, in order to limit the response burden or due to budgetary constraints.

 

	Target population when sample/census survey is used for collection of raw data	Target population when administrative data or pre-compiled statistics are used
Definition of the national target population	Data collection is based on registers of enterprises with known or assumed R&D activities, plus a stratified random sample of the remaining firms
Estimation of the target population size	Our target population consists of the register of enterprises with known or assumed R&D activities plus the population of enterprises not in this register from which the stratified random sample is drawn
Size cut-off point	Enterprises with 0 persons employed are excluded from sampling.
Size classes covered (and if different for some industries/services)	Micro firms with 1-9 persons employed are included in the census share of known or assumed R&D performers. In Brussels and Wallonia they are also included in all NACE classes included in the random sample share. In Flanders, micro firms were only included in the random sample share for NACE 19-23, 26-30, 59-63 and 71.
NACE/ISIC classes covered	All market oriented NACE codes are covered in the census share. For the random sample share: - Brussels: NACE 10-11, 13-17, 19-53, 58-62, 64-66, 69-74, 79-82; - Flanders: NACE 01-46, 49-53, 58-66, 69-74, 78-82, 95; - Wallonia: NACE 01-11, 13-17, 19-53, 58-62, 64-66, 69-74, 79-82.

3.6.2. Frame population – Description

The target population is the population for which inferences are made. The frame (or frames, as sometimes several frames are used) is a device that permits access to population units. The frame population is the set of population units which can be accessed through the frame and the survey data really refer to this population.

 

Method used to define the frame population	The Belgian official business register was used as frame population.
Methods and data sources used for identifying a unit as known or supposed R&D performer	A register of known or supposed R&D performers (either permanent or occasional R&D) is kept and updated using the following sources of information: - Lists of enterprises reporting R&D activities in previous R&D surveys, in innovation surveys (lists from regional governments authorities). - Lists of enterprises receiving regional governments grants and contracts for innovation and/or R&D during the surveyed reference period (lists from regional government authorities). - Lists of enterprises receiving European public funding for R&D (list from European government) - Lists of enterprises applying for certain tax incentives for R&D during the surveyed reference period (list owned by national government). - List of firms that include R&D numbers (above a certain threshold) in their annual accounts for the surveyed reference period (public data, commercialized in database format by a private company) - List of firms that reported import/export of R&D in Balance of Payments survey (list from the Belgian National Bank) - Other sources: members of fairly high-tech sector associations (biotech), list of academic spinoffs (owned by university technology transfer office), list of firms applying for patents, lists of firms cooperating with collective research centers, press articles and updates from the general media,...
Inclusion of units that primarily do not belong to the frame population	Non-profits and non-market oriented entities are excluded from the frame population.
Frequency and the methods applied for inclusion R&D performers not known and not supposed to perform R&D	A stratified random sample of the remaining population of firms allows identifying R&D performers not included in the register of known or supposed R&D performers. This is done each time the R&D survey is conducted, which is every two years (every even calendar year).
Number of “new”1) R&D enterprises that have been identified and included in the target population	RD performers newly detected in RD 2024, either from the random sample or from one of the external sources we use to detect potentially R&D active firms, e.g., list of firms with R&D grants/fiscal incentives, etc. (see list above): 1104 legal units 1018 enterprise units
Systematic exclusion of units from the process of updating the target population	For the census share of firms known or assumed to perform R&D activities, only entities without persons employed were excluded. There was no limitation in terms of NACE, as long as the economic activity was market oriented. The stratified random sampling process of firms taken outside the set of known or assumed R&D performers was limited to specific size classes and NACE categories. For more details, see section 3.6.1 above.
Estimation of the frame population	The frame population for reference year 2023 consisted of 1 032 350 market oriented enterprises

1)       i.e. enterprises previously not known or not supposed to perform R&D

3.7. Reference area

Not requested. R&D statistics cover national and regional data.

3.8. Coverage - Time

Not requested, see concept 12.3.3. (data availability).

3.9. Base period

The base year for the unit Purchasing Power Standard (PPS) and PPS per inhabitant at constant prices is currently 2005. All calculations of non-basic unit (national currencies) are done by Eurostat.

R&D expenditure is published in the following units:  Euro (MIO_EUR) and Euro per inhabitant (EUR_HAB); data are available in the following units: basic unit National currency (MIO_NAC); Purchasing Power Standard (MIO_PPS); Purchasing Power Standard at 2005 prices (MIO_PPS_KP05); Purchasing Power Standard per inhabitant at constant 2005 prices (PPS_HAB_KP05); Percentage of gross domestic product (PC_GDP); and Percentage of total R&D expenditure (PC_TOT - for the breakdown by source of funds).

R&D personnel data are published in full-time equivalent (FTE), in head count (HC), as a percentage of total employment and as a percentage of active population.

Calendar year 2023.

6.1. Institutional Mandate - legal acts and other agreements

See below.

6.1.1. European legislation

Legal acts / agreements:

Since the beginning of 2021, the collection of R&D statistics is based on the Commission Implementing Regulation (EU) No 2020/1197  of 30 July 2020 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistics. Regulation No 2020/1197 sets the framework for the collection of R&D statistics and specifies the main variables of interest and their breakdowns at predefined level of detail. The transmission of R&D data is mandatory for Member States and EEA countries.

6.1.2. National legislation

Existence of R&D specific statistical legislation	There is a legal agreement on how the three regions and the national government in Belgium organize the production of official STI statistics: Legal Agreement on Statsitics on R&D
Are respondents obliged by the national law to provide raw and administrative data:	No, the survey is voluntary for respondents.

6.1.3. Standards and manuals

• Frascati Manual 2015, Guidelines for Collecting and Reporting Data on Research and Experimental Development
• EBS Methodological Manual on R&D Statistics

6.2. Institutional Mandate - data sharing

Not requested.

7.1. Confidentiality - policy

A property of data indicating the extent to which their unauthorised disclosure could be prejudicial or harmful to the interest of the source or other relevant parties.

At the level of the ESS,  the EU regulation 223/2009 on European statistics defines confidential data as data which allows statistical units (respondents) to be identified, either directly - by formal identifiers such as respondents’ names, addresses, identification numbers - or indirectly - by using a combination of variables or characteristics such as age, gender, education - thereby disclosing individual information (see Article 2(1)(e) of regulation 223/2009). 

At national level:

• Confidentiality protection required by law: The Belgian Interfederal Institute of Statistics (IIS) coordinates the statistics production at the regional and national level in Belgium. It abides by the European Statistics Code of Practice, including its principle 5, on statistical confidentiality and data protection: Statistical Confidentiality Policy
• Confidentiality commitments of survey staff: All staff members of the regional and national offices are subjected to national statistical law. Data protection officers supervise the correct application of all relevant legal obligations.

7.2. Confidentiality - data treatment

Data cells compiled with data of less than 5 units are flagged as confidential. In case the value of a cell corresponds for 80% or more with an observed response, these cells are also flagged as confidential. Secondary confidentiality is also guarded.

8.1. Release calendar

Flanders:

• Core results documenting the extent to which Flanders approaches the 3% R&D target are published on June 30;
• A more detailed report is published on September 30.

At the national level: data are not released at a fixed date. We try to publish them in July or August after the official transmission to Eurostat.

8.2. Release calendar access

At Eurostat level this is: Release calendar - Eurostat (europa.eu)

8.3. Release policy - user access

In Flanders there are strict guidelines for advance access to the core results published on June 30.  Advance access is limited in time (max. 24 hours in advance), and is made publicly known.  During the allotted advance access users are not allowed to publicly announce results.

Once results in Flanders are published on their dedicated websites, they are available to the general public.

Although the national level is not subjected to this policy, we explicitly ask all data producing organisations to respect the official transmission to Eurostat, which is on the same day (June 30). After this transmission all concerned parties are free to disclose aggregated data on their websites.

Results based on the R&D survey are published biennially, in uneven years.
In even years Flanders publishes a 3% note with core R&D numbers (R&D expenditure, R&D personnel) on June 30 based on responses obtained in the innovation survey (CIS).

10.1. Dissemination format - News release

Please see the sub-concepts 10.1 to 10.5 in the full metadata view.

10.1.1. Availability of the releases

	Availability (Y/N)1)	Links
Regular releases	Y	Done by national and regional authorities. - Regional newsletters announce the release of the following statistics in Flanders: on June, 30th: R&D expenditure: Statistiek Vlaanderen R&D personnel: Statistiek Vlaanderen 3% note: Main R&D Indicators on September 30th: Full report on R&D numbers: Vlaams Indicatorenboek (chapters 2, 3.4, 3.5 and 3.6).   National data are published on the website of the Belgian Science Policy Office:  Main R&D Indicators  (BELSPO – MERI). Publication takes place after transmission to Eurostat
Ad-hoc releases	N

1) Y - Yes, N – No

10.2. Dissemination format - Publications

See below.

10.2.1. Availability of means of dissemination

Means of dissemination	Availability (Y/N)1)	Links
General publication/article	Y	For Belgium: Report on STI statistics, key figures, study series and booklet by Belgian Science Policy Office and by regional government authorities. For Flanders:  June 30th: Vlaanderen in cijfers 3% Nota September 30th: Full report on R&D numbers: Vlaams Indicatorenboek (chapters 2, 3.4, 3.5 and 3.6)
Specific paper publication (e.g. sectoral provided to enterprises)	N

1) Y – Yes, N - No

10.3. Dissemination format - online database

An online database is available on the BELSPO website: Meri Belspo Database

There are two ways to view the data: an interactive application allows the creation of user-defined graphs and tables with drop-down menus for the selection of the necessary variables;

A full data set with all available data is available for download under the heading complete data files in Excel format.

10.3.1. Data tables - consultations

Not requested.

10.4. Dissemination format - microdata access

As Eurostat receives no R&D micro-data from the reporting countries, users should contact directly the respective national statistical institute (NSI) for access to the micro-data.

10.4.1. Provisions affecting the access

Access rights to micro-data	A legal agreement was established granting the Belgian National Bank access to a specific subset of R&D microdata, to be used in national accounts (to calculate GDP), the balance of payments and FATS. Academic researchers need to apply for access to the microdata. They need to sign a confidentiality agreement. Access to the Belgian micro data files is only permitted following approval by a committee ;consisting of both national and regional representatives. For access to the regional data, researchers need to apply to the staff in charge of the regional statistics production.
Access cost policy	No costs are charged for consulting the microdata.
Micro-data anonymisation rules	Generally, no anonymisation procedures are applied to the microdata, except for (in some cases) leaving out the ID identifying each firm.

10.5. Dissemination format - other

See below.

10.5.1. Metadata - consultations

Not requested.

10.5.2. Availability of other dissemination means

Dissemination means	Availability (Y/N)1)	Micro-data / Aggregate figures	Comments
Internet: main results available on the national statistical authority’s website	Y	Aggregate figures	Electronic publishing (website) and paper publications (reports). The paper publications mentioned are also made available on the website of the public agency responsible for the surveys (as mentioned elsewhere, the R&D survey in Belgium is not conducted by the national statistical office but is the result of a cooperation between the Belgian Science Policy Office and regional statistics producers).
Data prepared for individual ad hoc requests	Y	Aggregate figures	Requests from different governments and public agencies (at the national or at the regional level), academic researchers and students, industry associations.
Other	N

1) Y – Yes, N - No

10.6. Documentation on methodology

We do not provide methodological documentation but we provide ad-hoc answers to users' requests.

10.6.1. Metadata completeness - rate

Not requested.

10.7. Quality management - documentation

Please see the sub-concept 10.7.1 in the full metadata view.

10.7.1. Documentation and users’ requests

Type(s) of data accompanying information available (metadata, graphs, quality reports, etc.)	Most of the paper or electronic publications contain a methodology section or definitions of the R&D variables. A note regarding the methodological changes in the R&D survey introduced in 2021 is also available on the website of the Belgian Science Policy Office.
Requests on further clarification, most problematic issues	We generally get no requests for further clarifications (except for further validation of increases or decreases in the numbers), just ad hoc requests for specific statistics.

11.1. Quality assurance

At Eurostat level, the common quality framework of the European Statistical System (ESS) is composed of the European Statistics Code of Practice, the Quality Assurance Framework of the ESS, and the general quality management principles (such as continuous interaction with users, continuous improvement, integration, and harmonisation).

We abide by the principles enumerated in the Qulaity Assurance Framework of the European Statistical System (Quality Assurance Framework).

11.2. Quality management - assessment

See our comments in the Section on accuracy. In the Flanders region university students are recruited and trained to follow up non-responding firms by phone, to motivate them to respond. In the data editing stage special attention is paid to the top R&D performers, given the skewed distribution of R&D expenditure and the impact these top performers have on the overall numbers. Even these top performers oftentimes need to be re-contacted to clarify inconsistencies in their responses. A similar methodology is applied in the Brussels Capitol Region and the Walloon Region but to a smaller extent because of a more limited availability of human resources (no students).

The R&D survey is voluntary in Belgium, which implies that respondents may not be too motivated to provide accurate, thoughtful responses, especially when it comes to consistency over time. We have been trying to make the survey mandatory, but, as this requires legal work and approval of employers' associations, this proves to be a difficult task.

Since 2005 Belgium has a tax incentive that reduces the wages for R&;D employees when certain criteria are met. The introduction of this tax incentive had an impact on the willingness of some firms to respond, for example to define their work as R&D. A drawback, however, was that some respondents tend now to limit their numbers for R&D personnel to those employees for whom they obtained the fiscal incentive.

As in most surveys, the treatment of non-responses as well as the check of the quality of the information given by the enterprises answering the questionnaire are major issues in our data processing.

In many enterprises, R&D figures are not directly available, and even when available they might not be in conformity with the prescriptions of the Frascati Manual. So, despite guidelines, definitions and precisions given in the R&D questionnaire, and despite an extensive set of editing controls (to check the coherence and quality of the data, e.g. during the online coding of the data), we are well aware quality issues are an ongoing concern.

12.1. Relevance - User Needs

Please see the sub-concept 12.1.1 in the full metadata view.

12.1.1. Needs at national level

Users’ class1)	Description of users	Users’ needs
Institutions	Federal (national) government and regional authorities Federal (national) and regional ministries of economy, innovation and research Statistical and policy analysis public agencies at the federal (national) level and at the regional level National Bank of Belgium (NBB)	Detailed statistics, analyses, specific data, focus on key economic or policy questions; NBB has access to a subset of microdata for use in its national accounts, balance of payments and FATS. The R&D microdata are also merged with fiscal microdata by a national government institute, to conduct policy evaluation studies.
Social actors	Employers’ associations Central economic council Private non-profit organisations	Specific statistics, analyses, main results.
Media	Newspapers and news magazines	Press conference and newspaper articles on core results of R&D.
Universities	Academic researchers and students	Detailed statistics, analyses, various research projects, raw (anonymized) data (subset of variables), micro data linked with other subsets of data.

1)       Users' class codification

1- Institutions:

• European level: Commission (DGs, Secretariat General), Council, European Parliament, ECB, other European agencies etc.
• in Member States, at the national or regional level: Ministries of Economy or Finance, other ministries (for sectoral comparisons), National Statistical Institutes and other statistical agencies (norms, training, etc.), and
• International organisations: OECD, UN, IMF, ILO, etc.

2- Social actors: Employers’ associations, trade unions, lobbies, among others, at the European, national or regional level.

3- Media: International or regional media – specialized or for the general public – interested both in figures and analyses or comments. The media are the main channels of statistics to the general public.

4- Researchers and students (Researchers and students need statistics, analyses, ad hoc services, access to specific data.)

5- Enterprises or businesses (Either for their own market analysis, their marketing strategy (large enterprises) or because they offer consultancy services)

6- Other (User class defined for national purposes, different from the previous classes. )

12.2. Relevance - User Satisfaction

To evaluate if users' needs have been satisfied, the best way is to use user satisfaction surveys.

12.2.1. National Surveys and feedback

Conduction of a user satisfaction survey or any other type of monitoring user satisfaction	No
User satisfaction survey specific for R&D statistics	We don't have a specific satisfaction survey
Short description of the feedback received	Not applicable

12.3. Completeness

Please see the sub-concept 12.3.2 in the full metadata view.

12.3.1. Data completeness - rate

All mandatory datastes transmitted. 

12.3.2. Completeness - overview

Completeness is assessed via comparison of the data delivered against the requirements of Commission Implementing Regulation (EU) No 2020/1197.

	Reasons for missing cells
Preliminary variables	Not applicable
Obligatory data on R&D expenditure	Not applicable
Optional data on R&D expenditure	Not applicable
Obligatory data on R&D personnel	Not applicable
Optional data on R&D personnel	We do not ask for the breakdown of other R&D personnel into either technicians and equivalent staff, or other supporting staff. We do not ask for the breakdown of qualification by gender. We do not ask for age nor for citizenship.
Regional data on R&D expenditure and R&D personnel	NUTS 2 results by performance sector are confidential.

12.3.3. Data availability

See below.

12.3.3.1. Data availability - R&D Expenditure

	Availability1)	Frequency of data collection	Gap years – years with missing data	Changes - Description	Changes - Year of introduction	Changes - Reasons
Source of funds	Y - 1993	before 2013: annual, since 2013: every 2 years.
Type of R&D	Y - 1993	before 2013: annual, since 2013: every 2 years
Type of costs	Y - 1993	before 2013: annual, since 2013: every 2 years
Socioeconomic objective	N
Region	Y - 1993	annual
FORD	N
Type of institution	Y - 2002	every 2 years		R&D activity of enterprises belonging to a foreign and/or multinational group, or to private vs. public enterprises

1) Y-start year, N – data not available

12.3.3.2. Data availability - R&D Personnel (HC)

	Availability1)	Frequency of data collection	Gap years – years with missing data	Changes - Description	Changes - Year of introduction	Changes - Reasons
Sex	Y-1998	before 2011: annual, since 2011: every two years
Function	Y-1998	before 2011: annual, since 2011: every two years
Qualification	Y-1998	before 2011: annual, since 2011: every two years
Age	N
Citizenship	N
Region	Y-1998	before 2011: annual, since 2011: every two years
FORD	N
Type of institution	N
Economic activity	Y-1998	Every 2 years
Product field	N
Employment size class	Y-2002-2011, dropped after 2011

1) Y-start year, N – data not available

12.3.3.3. Data availability - R&D Personnel (FTE)

	Availability1)	Frequency of data collection	Gap years – years with missing data	Changes - Description	Changes - Year of introduction	Changes - Reasons
Sex	Y-1998, dropped after 2011	annual
Function	Y -1993	before 2013: annual, since 2013: every two years
Qualification	Y -1993	before 2013: annual, since 2013: every two years
Age	N
Citizenship	N
Region	Y - 1993	annual
FORD	N
Type of institution	N
Economic activity	Y - 1993	annual, since 2019 every 2 years
Product field	N
Employment size class	Y - 2002	annual, since 2019 every 2 years

1) Y-start year, N – data not available

12.3.3.4. Data availability - other

Additional dimension/variable available at national level1)	Availability2)	Frequency of data collection	Breakdown variables	Combinations of breakdown variables	Level of detail
Performing sectors for Extramural expenditures	Y	every two years			overall total
Economic activity for Extramural expenditures	Y	every two years			overall total
Pilot: Number of R&D active enterprises	Y (confidential)	every two years
Pilot: internal vs. external funds	Y	every two years			overall total
Pilot: top 5/10/20/50/100 R&D performers	Y	every two years
Pilot: R&D by foreign-controlled vs. domestic enterprises; by multinationals	Y	every two years			overall total
Pilot: internal vs. external R&D personnel	Y	every two years			overall total

1) This question is optional. It refers to variables and breakdowns NOT asked by the Commission Implementing Regulation (EU) No 2020/1197 (neither as 'optional'), if R&D data for BES are collected for additional breakdowns or/and at more detailed level than requested.

2) Y-start year

12.3.3.5. R&D personnel - Cross-classification by function and qualification (if available in FTE and HC)

Cross-classification	Unit	Frequency
Not available

13.1. Accuracy - overall

Accuracy in the statistical sense denotes the closeness of computations or estimates to the exact or true values. Statistics are not equal with the true values because of variability (the statistics change from implementation to implementation of the survey due to random effects) and bias (the average of the possible values of the statistics from implementation to implementation is not equal to the true value due to systematic effects).

Several types of statistical errors occur during the survey process. The following typology of errors has been adopted:

1. Sampling errors. These only affect sample surveys. They are due to the fact that only a subset of the population, usually randomly selected, is enumerated.
2. Non-sampling errors. Non-sampling errors affect sample surveys and complete enumerations alike and comprise:

• • Coverage errors,
  • Measurement errors,
  • Non response errors and
  • Processing errors.

Model assumption errors should be treated under the heading of the respective error they are trying to reduce.

13.1.1. Accuracy - Overall by 'Types of Error'

	Sampling errors1)	Non-sampling errors1)				Model-assumption Errors1)	Perceived direction of the error2)
		Coverage errors	Measurement errors	Processing errors	Non- response errors
Total intramural R&D expenditure	4	5	2	3	1		+/-
Total R&D personnel in FTE	4	5	2	3	1		+/-
Researchers in FTE	4	5	2	3	1		+/-

1)  Ranking of the type(s) of errors that result in over/under-estimation, from the most important source of error (1) to the least important source of error (6). If errors of a particular type do not exist, the sign ‘:‘ is used.

2)  The perceived direction of the ‘overall’ error using the signs “+” for over estimation, “-” for under estimation and “+/-” when assumption of the direction of the error cannot be made for R&D.

13.1.2. Assessment of the accuracy with regard to the main indicators

Indicators	5 (Very Good)1)	4 (Good)2)	3 (Satisfactory)3)	2 (Poor)4)	1 (Very poor)5)
Total intramural R&D expenditure			X
Total R&D personnel in FTE			X
Researchers in FTE				X

1) 'Very Good' = High level of coverage (annual rate of substitution in the target population lower than 5%). High average rates of response (>80%) in census and sample surveys (BES R&D). Full data consistency with reference to totals and relationships between variables in the dataset sent to Eurostat.  

2) 'Good' = If at least one out of the three criteria described above is not fully met.

3) 'Satisfactory' = If the average rate of response is lower than 60% even by meeting the two remaining criteria.

4) 'Poor' = If the average rate of response is lower than 60% and at least one of the two remaining criteria is not met.

5) 'Very Poor' = If all the three criteria are not met.

13.2. Sampling error

That part of the difference between a population value and an estimate thereof, derived from a random sample, which is due to the fact that only a subset of the population is enumerated.

13.2.1. Sampling error - indicators

See below. 

13.2.1.1. Variance Estimation Method

None - we did not attempt to estimate variances and coefficients of variations (as by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises).

13.2.1.2. Confidence interval for key variables by NACE

	Industry sector1)	Services sector2)	TOTAL
R&D expenditure	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)
R&D personnel (FTE)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)

1)        Industry sector (NACE Rev. 2: 01-03, 05-09, 10-33, 35, 36-39, 41-43)

2)        Services sector (NACE Rev 2.: 45-47, 49-53, 55-56, 58-63, 64-66, 68, 69-75, 77-82, 84, 85, 86-88, 90-93, 94-96, 97-98, 99)

13.2.1.3. Confidence interval for key variables by Size Class

	0-9 employees and self-employed persons (optional)	10-49 employees and self-employed persons	50-249 employees and self-employed persons	250- and more employees and self-employed persons	TOTAL
R&D expenditure	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)
R&D personnel (FTE)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)	Not relevant (by far, the majority of R&D expenditure and R&D personnel come from the census share of our surveyed set of enterprises)

13.3. Non-sampling error

Non-sampling errors occur in all phases of a survey. They add to the sampling errors (if present) and contribute to decreasing overall accuracy. It is important to assess their relative weight in the total error and devote appropriate resources for their control and assessment.

13.3.1. Coverage error

Coverage errors (or frame errors) are due to divergences between the target population and the frame population. The frame population is the set of target population members that has a chance to be selected into the survey sample. It is a listing of all items in the population from which the sample is drawn that contains contact details as well as sufficient information to perform stratification and sampling.

• Description/assessment of coverage errors: Only 2.9% of the initial gross sample selected turned out to be non-eligible enterprises that had to be dropped from the net, post survey sample.
• Measures taken to reduce their effect: We use multiple sources of information to update the register of known or assumed R&D performers. As they are the major contributors to the final R&D numbers, we feel confident in our coverage of the target population. Moreover, we evaluate our sources for detecting newcomers in each survey wave. Newly detected enterprises generally only cover a limited share of R&D expenditure/personnel, implying our methods for updating our register of known or assumed R&D performers work well.

13.3.1.1. Over-coverage - rate

Not requested.

13.3.1.2. Common units - proportion

Not requested.

13.3.1.3. Frame misclassification rate

Misclassification rate measures the percentage of enterprises that changed stratum between the time the frame was last updated and the time the survey was carried out. It is defined as the number of enterprises that changed stratum divided by the number of enterprises which belong to the stratum, according to the frame. The rate can be estimated based on the characteristics of the surveyed enterprises.

 

By size class for the Industry Sector (NACE Rev. 2: 01-03, 05-09, 10-33, 35, 36-39, 41-43)	0-9 employees and self-employed persons (optional)	10-49 employees and self-employed persons	50-249 employees and self-employed persons	250-and more employees and self-employed persons	TOTAL
Number or surveyed enterprises in the stratum (according to frame)	631	1449	1167	404	3651
Number of surveyed enterprises that have changed stratum (after inspection of their characteristics)	5	43	80	37	165
Misclassification rate	1%	3%	7%	9%	5%
By size class for the Services Sector (NACE Rev 2.: 45-47, 49-53, 55-56, 58-63, 64-66 68, 69-75, 77-82, 84, 85, 86-88, 90-93, 94-96, 97-98, 99)	0-9 employees and self-employed persons (optional)	10-49 employees and self-employed persons	50-249 employees and self-employed persons	250-and more employees and self-employed persons	TOTAL
Number or surveyed enterprises in the stratum (according to frame)	2202	2064	1002	562	5830
Number of surveyed enterprises that have changed stratum (after inspection of their characteristics)	43	89	77	45	254
Misclassification rate	2%	4%	8%	8%	4%

13.3.2. Measurement error

Measurement errors occur during data collection and generate bias by recording values different than the true ones (e.g. difficulty to distinguish intramural from extramural R&D Expenditure). The survey questionnaire used for data collection may have led to the recording of wrong values, or there may be respondent or interviewer bias.

• Description/assessment of measurement errors:

We still are asked by firms whether the work they have been doing could be considered to be R&D or not. As there are many more firms that do not contact us, we can assume that many of them are still unsure as to what should be considered to be R&D or not.
Unity measure errors (e.g., cases where time spent on R&D is given in hours rather than FTE, or where expenditure is reported in millions rather than in euros) occur occasionally.
Group level responses are given occasionally, when unusually high numbers are given for R&D, given a firm’s basic economic data (turnover, number of persons employed).
Some respondents limit their numbers for R&D personnel to those for whom they obtained the tax incentive for R&D, yielding un underestimate of overall R&D personnel. Some firms had to return the tax incentives they initially had obtained for R&D personnel, definitely making them unwilling to respond to our R&D survey, even though their activities would still count towards R&D under Frascati Manual guidelines.
Several top R&D performers were unable to give head counts or FTE for consultants embedded in their own internal R&D projects, as they paid a flat fee for the work to be done (under their supervision). Several of these firms included consultants abroad. As those consultants they hired worked remotely, they were definitely not able to give more specifics such as gender and occupation of those consultants. Because of the difficulty in obtaining information on embedded consultants (“external R&D personnel” in the terminology of the 2015 Frascati Manual), we included in the statistics we reported for qualification of R&D personnel, only own, in-house R&D personnel, in line with what we have done in the past.

• Measures taken to reduce their effect:

We keep track of frequently made comments throughout the field phase, and try to take those comments into account when preparing the survey form for the next wave. Whenever we revise or add new questions, we perform cognitive interviews to test them.
We provide a diagram to try to help respondents in distinguishing between in-house and external R&D, and to decide on how consultants embedded in own R&D projects fit into the reporting.
A definition of what is R&D and what should be excluded is included in the survey form.
We provide help to respondents by phone, and they do often contact this helpdesk.
The online version of the questionnaire form contains logical consistency checks, that produce error messages/warnings in case of inconsistencies. The error messages are not binding.
We have an extensive data editing routine for detecting and correcting inconsistencies in survey responses. Especially for major R&D performers we try to re-contact respondents in order to clarify and correct their responses. We set up meetings with major R&D performers to make sure their responses are in line with Frascati Manual guidelines.

13.3.3. Non response error

Non-response occurs when a survey failed to collect data on all survey variables from all the population units designated for data collection in a sample or complete enumeration.

There are two elements of non-response:

• Unit non-response, which occurs when no data (or so little as to be unusable) are collected on a designated population unit.
• Item non-response, which occurs when data only on some, but not all survey variables are collected on a designated population unit.

The extent of response (and accordingly of non response) is also measured with response rates.

13.3.3.1. Unit non-response - rate

The main interest is to judge if the response from the target population was satisfying by computing the weighted and un-weighted response rate.

Definition: Eligible are the sample units which indeed belong to the target population. Frame imperfections always leave the possibility that some sampled units may not belong to the target population. Moreover, when there is no contact with sample units and no other way to establish their eligibility they are characterised as ‘unknown eligibility units’

Definition:

• Un-weighted Unit Non- Response Rate = [1 - (Number of units with a response) / (Total number of eligible and unknown eligibility units in the survey)] * 100 
• Weighted Unit Non- Response Rate = [1 - (Total weighted responding units) / (Total weighted number of eligible / unknown eligibility units in the sample)] * 100

13.3.3.1.1. Unit non-response rates by Size Class

	0-9 employees and self-employed persons (optional)	10-49 employees and self-employed persons	50-249 employees and self-employed persons	250-and more employees and self-employed persons	TOTAL
Number of units with a response in the realised sample	1713	2065	1372	712	5862
Total number of units in the sample	2950	3651	2331	1196	10128
Unit Non-response rate (un-weighted)	42%	43%	41%	40%	42%
Unit Non-response rate (weighted)	45%	53%	46%	38%	50%

13.3.3.1.2. Unit non-response rates by NACE

	Industry1)	Services2)	TOTAL
Number of units with a response in the realised sample	2386	3476	5862
Total number of units in the sample	4005	6123	10128
Unit Non-response rate (un-weighted)	40%	43%	42%
Unit Non-response rate (weighted)	50%	50%	50%

1)        Industry (NACE Rev. 2: 01-03, 05-09,10-33,35,36-39,41-43)

2)        Services (NACE Rev 2.: 45-47, 49-53, 55-56, 58-63, 64-66 68, 69-75, 77-82, 84, 85, 86-88, 90-93, 94-96, 97-98, 99)

13.3.3.1.3. Recalls/Reminders description

See section 18.3, “incentives used for increasing response” for a description of reminders sent.

13.3.3.1.4. Unit non-response survey

Conduction of a non-response survey	Due to limited resources, no non-response surveys were conducted.
Selection of the sample of non-respondents	Not applicable as no non-response  survey was conducted
Data collection method employed	Not applicable as no non-response  survey was conducted
Response rate of this type of survey	Not applicable as no non-response  survey was conducted
The main reasons of non-response identified	The questions are said (or seem) to be difficult, too specific or too detailed. The R&D data are reportedly not available in the enterprise or difficult to estimate. The questionnaire is too long and requires too many resources and too much time. Non R&D performers or small enterprises often times state the questionnaire is not relevant/does not apply to them. Confidentiality constraints are also  sometimes raised.

13.3.3.2. Item non-response - rate

Definition: Un-weighted Item non-Response Rate (%) = [1-(Number of units with a response for the item) / (Total number of eligible , for the item, units in the sample)] * 100

13.3.3.2.1. Un-weighted item non-response rate

	R&D Expenditure	R&D Personnel (FTE)	Researchers (FTE)
Item non-response rate (un-weighted) (%)	51%	54%	63%
Imputation (Y/N)	Yes: for register firms both in case of item non-response and unit non-response; and for sampled non-register firms only in case of item non-response.	Yes: for register firms both in case of item non-response and unit non-response; and for sampled non-register firms only in case of item non-response.	Yes: for register firms both in case of item non-response and unit non-response; and for sampled non-register firms only in case of item non-response.
If imputed, describe method used, mentioning which auxiliary information or stratification is used	Using ratio estimators for permanent R&D performers and conditional means for occasional R&D performers. Auxiliary variables used were, depending on availability: R&D expenditure or R&D personnel from previous two R&D surveys; turnover; total employment; R&D expenditure or R&D personnel from latest CIS.	Using ratio estimators for permanent R&D performers and conditional means for occasional R&D performers. Auxiliary variables used were, depending on availability: R&D expenditure or R&D personnel from previous two R&D surveys; turnover; total employment; R&D expenditure or R&D personnel from latest CIS.	Using ratio estimators for permanent R&D performers and conditional means for occasional R&D performers. Auxiliary variables used were, depending on availability: R&D expenditure or R&D personnel from previous two R&D surveys; turnover; total employment; R&D expenditure or R&D personnel from latest CIS.

13.3.3.3. Magnitude of errors due to non-response

	Magnitude of error (%) due to non-response
Total intramural R&D expenditure	Unknown
Total R&D personnel in FTE	Unknown
Researchers in FTE	Unknown

13.3.4. Processing error

Between data collection and the beginning of statistical analysis, data must undergo a certain processing: coding, data entry, data editing, imputation, etc. Errors introduced at these stages are called processing errors. Data editing identifies inconsistencies or errors in the data.

13.3.4.1. Identification of the main processing errors

Data entry method applied	Data keying: enterprises respond either on paper, or online (with a secure internet connection). If a paper questionnaire is returned, data entry is done with the electronic questionnaire so that the automatic controls (logical consistency checks) present in this format will apply. Each time an error is detected, a flag appears. Respondents are then able to correct inconsistencies while responding online. Remaining inconsistencies are handled in a separate data editing stage afterwards. In Flanders and Brussels, all paper responses are also entered twice in the online questionnaire format. Both entries are then compared to detect processing errors. Generally, several errors are found then, and are corrected by consulting the original paper response.
Estimates of data entry errors	Not available
Variables for which coding was performed	All nominal and ordinal variables are automatically recoded after data entry.
Estimates of coding errors	No errors
Editing process and method	Logical consistency checks are performed, plus comparisons to turnover and employment numbers whenever publicly available, and annual account data in general; checks for group level responses are performed (to avoid partial or double counting); comparisons are made to (edited) responses given to previous R&D and CIS surveys. Some editing rates that were obtained in Flanders: 2023 R&D expenditure: for 4% of the total number of reporting units that responded to have had either zero or non-zero R&D expenditure during the reference period (178 out of 4854 legal unit  responses), the original responses for R&D expenditure were modified during the data editing process, due to inconsistencies found in their responses. These modifications were mostly made within the subset of reporting units that had R&D activities during the reference period: within the subset of reporting units that had at least some R&D activities during the reference period, 5% of the original responses for R&D expenditure were modified during the data editing process, due to inconsistencies found in their responses (158 out of  3195 responses). 2023 Total R&D personnel as FTE: for 5% of the reporting units that responded to have had either no or some (internal or external) R&D personnel during the reference period (245 out of 4845 legal unit responses) the original responses for R&D personnel were modified, due to inconsistencies in their responses. Mostly these were cases with R&D activities. Within the subset of reporting units with at least some (internal or external) R&D personnel, this constituted 8% of this subset (243 out of 3195 responses) 2023 total researchers as FTE: for 11% of the reporting units that had some (internal or external) R&D personnel and that responded to the question asking for the breakdown of total R&D personnel over researchers and other technical or support staff, their original response to the question on number of researchers as FTE, was modified due to inconsistencies in their response (216 out of 1905 responses).
Procedure used to correct errors	Re-contact enterprise that has provided information, checks against economic or other variables whenever publicly available (annual accounts, web sites), imputation if no information from enterprise or other sources.

13.3.5. Model assumption error

Not requested.

14.1. Timeliness

Timeliness and punctuality refer to time and dates, but in a different manner: the timeliness of statistics reflects the length of time between their availability and the event or phenomenon they describe. Punctuality refers to the time lag between the release date of the data and the target date on which they should have been delivered, with reference to dates announced in the official release calendar.

14.1.1. Time lag - first result

Time lag between the end of reference period and the release date of the results:
Indicator: (Release date of provisional/ first results) - (Date of reference for the data)

• End of reference period: 31 December 2023
• Date of first release of national data: 31 October 2024
• Lag (days): 330 

14.1.2. Time lag - final result

• End of reference period: 31 December 2023
• Date of first release of national data: 30 June 2025
• Lag (days): 545

14.2. Punctuality

Punctuality refers to the time lag between the release date of data and the target date on which they were scheduled for release as announced officially.

14.2.1. Punctuality - delivery and publication

Punctuality of time schedule of data release = (Actual date of the data release) - (Scheduled date of the data release).

14.2.1.1. Deadline and date of data transmission

	Transmission of provisional data	Transmission of final data
Legally defined deadline of data transmission (T+_ months)	10	18
Actual date of transmission of the data (T+x months)	10	20
Delay (days)	0	36
Reasoning for delay	Not applicable	Transition to new platform to prepare the data for transmission; increase of data requirements: significantly more data cells need to be transmitted due to the breakdown of total R&D personnel into internal and external R&D personnel

15.1. Comparability - geographical

This sub-concept refers to the geographical comparability of data among the 27 Member States and the EFTA and Candidate Countries.

15.1.1. Asymmetry for mirror flow statistics - coefficient

Not requested.

15.1.2. General issues of comparability

Not applicable

15.1.3. Survey Concepts Issues

The following table lists a number of key survey concepts and conceptual issues; it gives reference to the Commission Implementing Regulation (EU) No 2020/1197 or Frascati manual (FM) and EBS Methodological Manual on R&D Statistics paragraphs with recommendations about these concepts / issues.

 

Concept / Issues	Reference to recommendations	Deviation from recommendations	Comments on national definition / Treatment – deviations from recommendations
R&D personnel	FM2015 Chapter 5 (mainly sub-chapter 5.2).	No deviation
Researcher	FM2015, §5.35-5.39.	No deviation
Approach to obtaining Headcount (HC) data	FM2015, §5.58-5.61 (in combination with Eurostat's EBS Methodological Manual on R&D Statistics).	No deviation
Approach to obtaining Full-time equivalence (FTE) data	FM2015, §5.49-5.57 (in combination with Eurostat’s EBS Methodological Manual on R&D Statistics).	No deviation	The criterion of at least 10% of time spent on R&D was applied at the aggregate level, not at the level of individual firms.
Reporting data according to formula: Total R&D personnel = Internal R&D personnel + External R&D personnel	FM2015, §5.25	Yes
Intramural R&D expenditure	FM2015 Chapter 4 (mainly sub-chapter 4.2).	No deviation
Special treatment for NACE 72 enterprises	FM2015, § 7.59.	No deviation
Statistical unit	FM2015 Chapter 7 (mainly paragraphs 7.3 and 7.7 in combination with Eurostat's  EBS Methodological Manual on R&D Statistics).	No deviation
Target population	FM2015 Chapter 7 (mainly sub-chapter 7.7 in combination with Eurostat's EBS Methodological Manual on R&D Statistics).	No deviation
Identification of not known R&D performing or supposed to perform R&D enterprises	FM2015 Chapter 7 (mainly sub-chapter 7.7 in combination with Eurostat's  EBS Methodological Manual on R&D Statistics)	No deviation
Sector coverage	FM2015 Chapter 3 (mainly sub-chapter 3.5) in combination with Eurostat's EBS Methodological Manual on R&D Statistics).	No deviation
NACE coverage and breakdown	Reg. 2020/1197 : Annex 1, Table 18	No deviation
Enterprise size coverage and breakdown	Reg. 2020/1197 : Annex 1, Table 18	No deviation
Reference period for the main data	Reg. 2020/1197 : Annex 1, Table 18	No deviation
Reference period for all data	Reg. 2020/1197 : Annex 1, Table 18	No deviation

 

15.1.4. Deviations from recommendations

The following table lists a number of key methodological issues, which may affect the international comparability of national R&D statistics. The table gives the references in the Frascati manual (FM), where related recommendations are made. Countries are asked to report on the existence of any deviations from existing recommendations and comment upon.

 

Methodological issues	Reference to recommendations	Deviation from recommendations	Comments on national treatment / treatment deviations from recommendations
Data collection preparation activities	FM2015 Chapter 7 (mainly sub-chapter 7.7).	No deviation
Data collection method	FM2015 Chapter 7 (mainly sub-chapter 7.7).	No deviation	Reporting units are legal units; statistical units used in statistics transmitted are enterprises. Regional data are based on the establishment units within enterprises where the majority of the enterprise’s R&D occurs.
Cooperation with respondents	FM2015 Chapter 7 (mainly sub-chapter 7.7).	No deviation
Follow-up of non-respondents	FM2015 Chapter 7 (mainly sub-chapter 7.7).	No deviation
Data processing methods	FM2015 Chapter 6 (mainly sub-chapter 6.7).	No deviation
Treatment of non-response	FM2015 Chapter 6 (mainly sub-chapter 6.7).	No deviation
Data weighting	FM2015 Chapter 7 (mainly sub-chapter 7.7).	No deviation
Variance estimation	FM2015 Chapter 6 (mainly sub-chapter 6.9).
Data compilation of final and preliminary data	Reg. 2020/1197 : Annex 1, Table 18	No deviation
Survey type	FM2015 Chapter 6 (mainly sub-chapter 6.4).	No deviation
Sample design	FM2015 Chapter 6 (mainly sub-chapter 6.4).	No deviation
Survey questionnaire	FM2015 Chapter 6 (mainly sub-chapter 6.4).	No deviation	Multiple languages (Dutch, French, English, and German); short and long form; regionalized versions; special versions for units in Nace72 (R&D) sector and group answers in Flanders . 16 different forms in total.

15.2. Comparability - over time

For more information related to the break years and the nature of the breaks, see the following sub-concepts in the full metadata view.

15.2.1. Length of comparable time series

See below.

15.2.2. Breaks in time series

	Length of comparable time series	Break years1	Nature of the breaks
R&D personnel (HC)		2021, 1990, 1981	2021: for the first time total R&D personnel = sum of internal and external R&D personnel; in all earlier years only internal R&D personnel was reported. 1990: R&D expenditure and personnel of centres for collective research controlled by and primarily serving business enterprises were underestimated in the period 1977-1989. 1981:R&D personnel in the business enterprise sector was underestimated given that no R&D personnel data for public enterprises were available. However, over the course of the 90s the public enterprises were included in the business enterprise R&D survey and this survey includes questions on R&D personnel.
Function		2021	2021: for the first time total R&D personnel = sum of internal and external R&D personnel; in all earlier years only internal R&D personnel was reported.
Qualification
R&D personnel (FTE)		2021, 1990, 1981	2021: for the first time total R&D personnel = sum of internal and external R&D personnel; in all earlier years only internal R&D personnel was reported. 1990: R&D expenditure and personnel of centres for collective research controlled by and primarily serving business enterprises were underestimated in the period 1977-1989. 1981:R&D personnel in the business enterprise sector was underestimated given that no R&D personnel data for public enterprises were available. However, over the course of the 90s the public enterprises were included in the business enterprise R&D survey and this survey includes questions on R&D personnel
Function		2021	2021: for the first time total R&D personnel = sum of internal and external R&D personnel; in all earlier years only internal R&D personnel was reported
Qualification
R&D expenditure		1992, 1990, 1989, 1988, 1987, 1983, 1981	1992: A new methodology for estimating aggregate BERD was introduced. The enterprises surveyed were taken from an inventory of enterprises known or assumed to conduct R&D on a permanent basis (these firms accounted for approximately 90% of aggregate BERD) plus a random sample representing the population of Belgian enterprises outside the register of "known" R&D performers. There was therefore a major break in the series, particularly with regard to the services sector which hitherto had been under-represented. The data for 1992 cannot be compared with those available for earlier years. 1990: R&D expenditure and personnel of centres for collective research controlled by and primarily serving business enterprises were underestimated for the period 1977-1989. 1989: The methods for evaluating R&D activities changed due a state reform. After the reform, production of R&D statistics was moved from the national to the regional level (NUTS1) by law. 1987 and 1988: Gross domestic expenditure on R&D (GERD) and GERD funded by the state (public authorities) were underestimated because R&D financed by the federal authorities was excluded (approximately 2-4% of GERD and 7-15% of GERD financed by the state). Since no breakdown is available for the distribution of these funds by sector of performance, it is impossible to make an accurate estimate of the impact on the other more detailed data by sector of performance for the business enterprise and higher education sectors, for which data had been underestimated. 1983: The data relating to GERD are no longer comparable with those from earlier years due to changes in the methods used to evaluate the R&D effort in Belgium. 1981: R&D expenditure by public enterprises was accounted for solely in aggregate BERD and was not broken down by sector of industrial activity.
Source of funds
Type of costs
Type of R&D
Other

1)       Breaks years are years for which data are not fully comparable to the previous period.

15.2.3. Collection of data in the even years

Collection of data for even reference years (when the R&D survey is off; see also section 18.5.2): in one of the 3 NUTS 1 regions, numbers for intramural R&D expenditure and R&D personnel as FTE are obtained from the innovation survey (CIS) that surveys that reference year. CIS includes detailed questions asking for these two aspects. For the other two regions intramural R&D expenditure and internal R&D personnel as FTE are obtained from autoregressive models AR(1) on numbers obtained from previous surveys at the region level (the sample starts in 1993). For external personnel  and researchers as FTE, the percentages of external personnel and of researchers obtained in the last survey are applied to the current number of total R&D personnel as FTE. A more precise description is provided in Section 18.5.2.

It should be noted, however, that the two regions that apply regressions to obtain R&D numbers for even reference years, survey two years in their regular R&D survey (t and t-1), so the estimates obtained from regressions provided for the even reference years are updated using numbers obtained from an actual survey at t + 30. Hence, at t + 30, updated R&D numbers are transmitted to EUROSTAT for even reference years, resulting in R&D numbers obtained from actual surveys for all 3 regions for those reference years.

15.3. Coherence - cross domain

This part deals with any national coherence assessments which may have been undertaken. It reports results for variables which are the same or relevant to R&D statistics, from other national surveys and / or administrative sources and explains and comments on their degree of agreement with R&D statistics.  Intramural R & D expenditure (code 230101 in the Commission Implementing Regulation (EU) 2020/1197) and R & D personnel (code 230201) are surveyed also in foreign-controlled EU enterprises statistics (inward FATS).

The Community innovation survey also collects the R&D expenditure of enterprises that form the coverage of the CIS survey.

15.3.1. Coherence - sub annual and annual statistics

Not requested.

15.3.2. Coherence - National Accounts

Not available

15.3.3. National Coherence Assessments

Variable name	R&D Statistics - Variable Value	Other national statistics - Variable value	Other national statistics - Source	Difference in values (of R&D statistics)	Explanation of / comments on difference
BERD 2022  - Total	12 642 214	11 432 896	CIS 2022	1 209 318	Differences in size and sector coverage (very small enterprises and some services not covered in CIS).
BERD 2022  - Manufacturing	6 047 489	5 507 197	CIS 2022	540 292	Differences in size coverage (very small enterprises not covered in CIS).
BERD 2022  - Services of the business economy	6 063 687	5 577 958	CIS 2022	485 729	Differences in size and sector coverage (“Innovation core services activities (Com.Reg. 995/2012)” in CIS vs. “Services of the business economy” in BERD).

15.4. Coherence - internal

Please see the sub-concepts 15.4.1 and 15.4.2 in the full metadata view.

15.4.1. Comparison between preliminary and final data

This part compares key R&D variables as preliminary and final data.

	Total R&D expenditure (in 1000 of national currency)	Total R&D personnel (in FTEs)	Total number of researchers (in FTEs)
Preliminary data (delivered at T+10)	14 676 150	89 649	52 007
Final data (delivered T+18)	14 095 469	79 281	47 718
Difference (of final data)	580 681	10 368	4 289

 

Comments :

....

15.4.2. Consistency between R&D personnel and expenditure

	Average remuneration per year (cost in national currency)	Explanation of consistency issues if any
Consistency between FTEs of internal R&D personnel and R&D labour costs (1)	Internal personnel costs (EUR):  6 789 330 733 Internal R&D personnel in FTE:  63 776 Average remuneration (EUR): 106 456
Consistency between FTEs of external R&D personnel and other current costs for external R&D personnel (2)	External R&D personnel in FTE   14 335 ternal personnel costs (EUR)         2 623 784 972 Average remuneration (EUR)        183 034

(1)    Calculate the average remuneration (cost) of individuals belonging to the internal R&D personnel, excluding those who are only formally ‘employees’ (university students, grant holders, etc.).

(2)    Calculate the average remuneration (cost) of individuals belonging to the external R&D personnel (FTEs/other current R&D costs for external R&D personnel).

The assessment of costs associated with a statistical product is a rather complicated task since there must exist a mechanism for appointing portions of shared costs (for instance shared IT resources and dissemination channels) and overheads (office space, utility bills etc). The assessment must become detailed and clear enough so that international comparisons among agencies of different structures are feasible.

16.1. Costs summary

	Costs for the statistical authority (in national currency)	Cost for the NSI in time use / person / day
Staff costs	Not separately available	The R&D survey is conducted in a decentralized way in Belgium, the 3 NUTS-1 regions are each responsible for their own data collection and processing.
Data collection costs	Not separately available	The R&D survey is conducted in a decentralized way in Belgium, the 3 NUTS-1 regions are each responsible for their own data collection and processing.
Other costs	Not separately available	The R&D survey is conducted in a decentralized way in Belgium, the 3 NUTS-1 regions are each responsible for their own data collection and processing.
Total costs	Not separately available	The R&D survey is conducted in a decentralized way in Belgium, the 3 NUTS-1 regions are each responsible for their own data collection and processing.

The shares of the figures given in the first column that are accounted for by payments to private firms or other Government agencies.

 

 

16.2. Components of burden and description of how these estimates were reached

	Value	Computation method
Number of Respondents (R)	5862 legal units, 5626 enterprises.	Count of valid unit responses
Average Time required to complete the questionnaire in hours (T)1	Unknown	Not asked in the survey forms
Average hourly cost (in national currency) of a respondent (C)	Unknown	Not asked in the survey forms
Total cost	Unknown	Not asked in the survey forms

1)        T = the time required to provide the information, including time spent assembling information prior to completing a form or taking part in interview and the time taken up by any subsequent contacts after receipt of the questionnaire (‘re-contact time’)

17.1. Data revision - policy

Not requested.

17.2. Data revision - practice

Not requested.

17.2.1. Data revision - average size

Not requested.

18.1. Source data

Several separate activities are used for the collection of raw data or pre-compiled administrative data and statistics related to R&D. This section collects information on the type of data collection instruments used as well as methodological information for each data collection instrument. Depending on the type of data collection instrument used, only the sections corresponding to that data collection instrument are filled in.

18.1.1. Data source – general information

Survey name	BES R&D survey
Type of survey	The survey is voluntary.
	The survey is a combination of a census (register of enterprises known to engage in R&D activity, either continuously or occasionally, or supposed to perform R&D) and of a stratified random sample to collect information on the R&D activities of enterprises which are not included in the register of known R&D performers. The stratified random sample is achieved at the regional level (NUTS1) in Belgium. The three Belgian Regions are: Flanders, Wallonia and Brussels.
	The distribution of total R&D expenditure between registers and sample is as follows: registers (census share) represent about of 99% of BERD (2023) and the additional sample hardly 1 %.
Combination of sample survey and census data	For details on the target population for the stratified random sample: see the section on statistical population.
	The register of “known or supposed” R&D performers covers all NACE Rev. 2 divisions, given that there are at least some employees (zero employee entities are excluded).
Combination of dedicated R&D and other survey(s)	No
Sub-population A (covered by sampling)	Yes, a stratified random sample is taken from the general population of firms outside of the register of known or assumed R&D performers.  More details are given elsewhere.
Sub-population B (covered by census)	The register of “known or supposed” R&D performers covers all firms with at least one employee and all market oriented NACE Rev. 2 divisions.
Variables the survey contributes to	All mandatory variables in Commission Implementing Regulation (EU) No 2020/1197 plus a small number of optional variables (qualification, extramural R&D expenditure).
Survey timetable-most recent implementation
	Winter 2023-2024: questionnaire design;
	December 2023 - February 2024: sampling;
	April 2024 - December 2024: field phase;
	November 2024 – March 2025: data editing;
	March 2025 – June 2025: data weighting, data imputation, calculation of statistics;
	June 2025 – September 2025: final reports;
	October 2025: quality report/SIMS; remaining statistics.

18.1.2. Sample/census survey information

Sampling unit	Enterprise
Stratification variables (if any - for sample surveys only)	Economic activities (NACE), size classes and R&D status.
Stratification variable classes	Economic activities (NACE), size class, and R&D status. The stratified random sample is achieved at the regional level in Belgium. The three Belgian Regions are: Flanders, Wallonia, and Brussels (NUTS1). See more details elsewhere
Population size	The total population size for Belgium is 23848 enterprises. These enterprises represent 24323 legal units, with the following regional repartition: Brussels: 2838 legal units Flanders: 14690 legal units Wallonia: 6795 legal units
Planned sample size	10128 legal units, 9649 enterprises
Sample selection mechanism (for sample surveys only)	Stratified random selection mechanism.
Survey frame	The sampling from the population remaining after exclusion of the register of known or assumed R&D performers,  is done using the official business register of the national statistical office, Statistics Belgium This official business register is both at the level of legal units and at the enterprise level.
Sample design	Our survey design is as follows: · register of known or supposed R&D performers: census. · enterprises with 250 employees or more in targeted NACE divisions outside the register: census. · enterprises in NACE division 72: census. · stratified random sampling is done at the regional level for the units outside the register of known or assumed R&D performers. Strata are made by combining size and aggregates of NACE sectors.Thresholds can differ by region: · Flanders: The sample size per cell was set at 21; all enterprises were taken if the cell contained fewer than 21 enterprises, yielding a total sample size of 1100 enterprises for the random sample share. NACE sectors considered were 01-46, 49-53, 58-66, 69-74, 78-82, 95.    · Brussels and Wallonia: Cells were sampled proportionally to their overall size.  The total sample size for the random sample share was set at 250 for Brussels, and at 500 for Wallonia.  NACE sectors considered were 10-11, 13-17, 19-53, 58-62, 64-66,, 69-74, 79-82 for Brussels, and 01-11, 13-17, 19-53, 58-62, 64-66, , 69-74, 79-82 for Wallonia. In Flanders 38 strata were considered for the random sample share taken from small and medium size enterprises; in Wallonia and Brussels the corresponding numbers were 38 and 36, respectively.
Sample size	The net sample consists of 10128 legal units and 9649 enterprises.
Survey frame quality	The sampling of the remaining population of firms which are not included in the register is done using the population of the national statistical office, Statistics Belgium. This official register is both at the legal unit level and at the enterprise level. Self-employed persons, non-profits, non-market oriented public enterprises, foreign enterprises with no representation in Belgium, etc. are removed before sampling.
Variables the survey contributes to	All mandatory variables plus a small number of optional variables

18.1.3. Information on collection of administrative data or of pre-compiled statistics

Source	Euro Groups Register (EGR), Bel-first (commercial database), firms’ balance sheets, information from the Belgian National Bank who provides FATS, SBS/official business register values for turnover and persons employed.
Description of collected data / statistics	foreign-controlled affiliate vs. domestic enterprise; multinational enterprise, size class..
Reference period, in relation to the variables the administrative source contributes to	EGR: April 2025 version; Bel-first: May-October 2025 version (Bel-first only displays the current group structure status; no back versions of the Bel-first database are available), ); SBS/official business register: May 2025
Variables the administrative source contributes to	MNE with foreign decision centre, MNE with domestic decision centre and domestic firms without affiliates abroad, size class.

18.2. Frequency of data collection

See 12.3.3.

18.3. Data collection

Please see the sub-concepts 18.3.1 and 18.3.2 in the full metadata view.

18.3.1. Data collection overview

Realised sample size (per stratum)	The overall realized sample is 5626 enterprises and 5862 legal units (reporting units) for Belgium: 419 enterprises and 420 legal units for Brussels  595 enterprises and 596 legal units for Wallonia 4627 enterprises and 4864 legal units for Flanders
Mode of data collection	Generally, we do our first mailing by sending paper survey forms to sampled enterprises. These questionnaire forms contain login codes and passwords which give users the possibility to complete the survey form online. Because of the non-mandatory nature of the survey, several reminders are sent to improve the reponse rate. In the final stage of the field phase some regions (Flanders en Brussels) use personalised e-mails to contact persons within the enterprises. Given the fact that R&D expenditure is highly concentrated within a relatively small group of enterprises, extra efforts are done to obtain the required information for these entities. Established R&D enterprises receive an extensive questionnaire form, entities in the stratified random sample share and certain enterprises expected to have only occasional R&D activities receive an abbreviated survey form. If entities become established R&D performers they will receive the longer survey form in the next survey. Since the COVID-19 crisis Flanders has adopted another approach: a transition was made from a predominantly paper survey to a predominantly online survey. E-mail addresses were collected for contacts for each firm to be surveyed. These were sent a survey invitation by e-mail. Surveyed firms had the option to request a paper survey form, but rarely did so. Several e-mail reminders were sent, and one paper reminder that included a paper survey booklet. Sampled firms for whom no e-mail address could be found, were invited to respond on paper twice. The Walloon and Brussels region followed the usual approach. In an attempt to improve the response rates in the Brussels region personalised e-mails were sent to large companies in the service sector.
Incentives used for increasing response	To decrease the response burden, firms in the stratified random sample share and certain firms expected to have only occasional R&D activities are sent an abbreviated survey form. Personalised correspondence is preferred whenever possible. In Flanders five reminders were sent, four by e-mail and one by postal mail. In addition, following the first reminder follow-up phone calls were made to encourage firms to respond. In Brussels initially one reminder was sent; personalised e-mails were sent to large companies in the service sector. In Wallonia three e-mail reminders were sent.
Follow-up of non-respondents	The R&D survey is not mandatory in Belgium, contrary to the legal obligation in most European member states. Therefore, follow-up phone calls are made to improve response rates. These calls are made to non-respondents and to entities providing partial reponses.
Replacement of non-respondents (e.g. if proxy interviewing is employed)	No
Response rate (ratio of completed "interviews" over total number of eligible enterprises or enterprises of unknown eligibility)	The overall (weighted) response rate is 50% for Belgium. For more information, see the Table in Section 13.3.3.1.1.
Non-response analysis (if applicable -- also see section 18.5.4 Data compilation - Weighting and Estimation methods)	A non-response analysis is not conducted, due to limited resources.

18.3.2. Questionnaire and other documents

Annex	Name of the file
R&D national questionnaire and explanatory notes in English:	Short questionnaire for Brussels and the Walloon Region (EN): RD_BESSI_A_BE_2023_0000_AN_1.docx   Long questionnaire for Brussels and the Walloon Region (EN): RD_BESSI_A_BE_2023_0000_AN_2.docx Long questionnaire for the Flanders Region (EN): RD_BESSI_A_BE_2023_0000_AN_8.docx
R&D national questionnaire and explanatory notes in the national language:	Long questionnaire for the Flanders Region (NL): RD_BESSI_A_BE_2023_0000_AN_3.docx Short questionnaire for the Flanders Region (NL): RD_BESSI_A_BE_2023_0000_AN_9.docx Short questionnaire for Brussels and the Walloon Region (FR): RD_BESSI_A_BE_2023_0000_AN_4.docx Short questionnaire for Brussels and the Walloon Region (NL): RD_BESSI_A_BE_2023_0000_AN_5.docx Long questionnaire for Brussels and the Walloon Region (FR): RD_BESSI_A_BE_2023_0000_AN_6.docx Long questionnaire for Brussels and the Walloon Region (NL): RD_BESSI_A_BE_2023_0000_AN_7.docx
Other relevant documentation of national methodology in English:	Not available
Other relevant documentation of national methodology in the national language:	Not available

Annexes:
Long questionnaire for Brussels and the Walloon Region (EN):
Short questionnaire for Brussels and the Walloon Region (EN)
Long questionnaire for the Flanders Region (NL)
Short questionnaire for Brussels and the Walloon Region (FR)
Short questionnaire for Brussels and the Walloon Region (NL)
Long questionnaire for Brussels and the Walloon Region (FR):
Long questionnaire for Brussels and the Walloon Region (NL):
Long questionnaire for the Flanders Region (EN)
Short questionnaire for the Flanders Region (NL)

18.4. Data validation

Data are validated by the 3 Regions participating in the R&D survey in Belgium.

18.5. Data compilation

See below.

18.5.1. Imputation - rate

Imputation is the method of creating plausible (but artificial) substitute values for all those missing.
Definition:
Imputation rate (for the variable x) % = (Number of imputed records for the variable x) * 100 / (Total number of possible records for x)

18.5.1.1. Imputation rate by Size class

Size class	R&D Expenditure		R&D personnel (FTE)
	Unweighted	Weighted	Unweighted	Weighted
0-9 employees and self-employed persons (optional)	53 %	49 %	57 %	43 %
10-49 employees and self-employed persons	52 %	48 %	57 %	46 %
50-249 employees and self-employed persons	49 %	37 %	52 %	43 %
250-and more employees and self-employed persons	47 %	33 %	50 %	34 %
TOTAL	52 %	36 %	55 %	39 %

18.5.1.2. Imputation rate by NACE

NACE	R&D Expenditure		R&D personnel (FTE)
	Unweighted	Weighted	Unweighted	Weighted
Industry1)	48 %	36 %	52 %	39 %
Services2)	54 %	37 %	57 %	40 %
TOTAL	52 %	36 %	55 %	39 %

 

1)       Industry (NACE Rev. 2: 01-03, 05-09,10-33,35,36-39,41-43)

2)       Services (NACE Rev 2.: 45-47, 49-53, 55-56, 58-63, 64-66 68, 69-75, 77-82, 84, 85, 86-88, 90-93, 94-96, 97-98, 99)

18.5.2. Data compilation methods

	Data compilation method - Final data	Core R&D numbers (R&D expenditure, R&D personnel as FTE, researchers as FTE) for even reference years: T+18: Flanders: the numbers for R&D expenditure and R&D personnel as FTE are derived from the Community Innovation Survey (CIS) that is conducted in the year the R&D survey is off. The number for researchers as FTE is estimated by applying the rate of researchers among overall R&D personnel obtained in the last R&D survey, to the R&D personnel numbers obtained in CIS. Brussels and Wallonia: numbers for R&D expenditure and internal R&D personnel as FTE are estimated using an autoregressive AR(1) model that is applied to the historical growth rates numbers. For researchers as FTE the percentage of researchers among the total R&D personnel that was obtained in the last survey within each region is applied to the total R&D personnel numbers of each region. The same applies to external personnel and external researchers, with the appropriate ratios for each variable. T + 30: In Flanders: data from the CIS survey are not updated In Brussels and Wallonia: The 2024 R&D survey in these two regions covered the two years preceding the year in which the survey was conducted (2024).The survey data that were obtained this way were used to update the numbers for Belgium as a whole for the 2022 reference year. Those updates were transmitted to EUROSTAT on June 30, 2025, at T+30.
	Data compilation method - Preliminary data	R&D expenditure and internal R&D personnel as FTE: A autoregressive AR(1) model is applied to the historical growth rate numbers for each NUTS 1 region separately. The samples for estimating the parameters for the AR(1) models span from 1993 to the most recent available data. Internal Researchers as FTE: the percentage of researchers among the internal R&D personnel that was obtained within each region in the most recent R&D survey, is applied to the internal R&D personnel numbers of each region to get an estimate of the internal R&D researchers. External personnel as FTE: the ratio of external over internal R&D personnel that was obtained within each region in the most recent R&D survey, is applied to the internal R&D personnel numbers of each region to get an estimate of external personnel External researchers as FTE: the percentage of researchers among external R&D personnel that was obtained within each region in the most recent R&D survey, is applied to the external R&D personnel numbers of each region to get an estimate of external researchers

18.5.3. Measurement issues

Method of derivation of regional data	The BES R&D survey contains a question asking at which local establishment unit of the firm legal unit the majority of its R&D is performed. This is used to derive regional data for enterprises at the NUTS 2 level.
Coefficients used for estimation of the R&D share of more general expenditure items	Not applicable for the BES R&D survey
Inclusion or exclusion of VAT and provisions for depreciation in the measurement of expenditures	VAT is supposed to be excluded

18.5.4. Weighting and estimation methods

Weight calculation method	Census share enterprises were given a weight of 1. Imputations were made for them in case of unit non-response. When there were fewer than 3 responding enterprises in a cell of the random sample share design, they were also given a weight of 1. Cells considered for the random sample share were NACE cells for which reporting to EUROSTAT was mandatory, crossed by size class and NUTS 1 region. For all cells of the random sample share design with 3 or more observations, the inverse sampling fraction was used as a starting point. Post-stratification was done to account for non-response in a second step. In a third step, calibration was done using the Calmar macro available for SAS. The truncated linear method was used. Calibration was done on number of enterprises, turnover, and number of persons employed, whenever feasible. For a majority of cells, however, calibration on turnover and number of persons employed was not feasible, due to the small number of observations they contained.
Data source used for deriving population totals (universe description)	Population totals for number of enterprises, turnover and number of persons employed are taken from SBS/the official business register maintained by Statistics Belgium.
Variables used for weighting	Number of enterprises, turnover and number of persons employed.
Calibration method and the software used	Calibration was done using the Calmar macro available for SAS. The truncated linear method was used. Calibration was done on number of enterprises, turnover, and number of persons employed, whenever feasible. For a majority of cells, however, calibration on turnover and number of persons employed was not feasible, due to the small number of observations they contained.
Estimation	See elsewhere in this report: imputation methods are used for estimating missing values on the basis of additional information such as the previous answer for the same enterprise or information from the same survey. (Using ratio estimators for permanent R&D performers and conditional means for occasional R&D performers. Auxiliary variables used, depending on availability are: R&D expenditure or R&D personnel from previous two R&D surveys; turnover; total employment; R&D expenditure or R&D personnel from latest CIS.)

18.6. Adjustment

Not requested.

18.6.1. Seasonal adjustment

Not requested.

Statistics on Business enterprise R&D (BERD) measure research and experimental development (R&D) performed in the business enterprise sector, i.e., R&D expenditure and R&D personnel. In line with this objective, the target population for the national R&D survey of the business enterprise sector consists of all R&D performing enterprises (including all R&D performers – occasional and continuous, known and unknown – in all branches and size classes) belonging to this sector. The “enterprise” is defined in Council Regulation (EEC) No 1993/696 of 15 March 1993. The results are related to the population of all R&D performing enterprises classified in Sections A to U of the common statistical classification of economic activities as established by Regulation (EC) No 1893/2006 of the European Parliament and of the Council (NACE Rev.2).

The main concepts and definitions used for the production of R&D statistics are given by OECD (2015), Frascati Manual, which is the internationally recognised standard methodology for collecting R&D statistics, and by Eurostat’s EBS Methodological Manual.

Since the beginning of 2021, the collection of R&D statistics is based on Commission Implementing Regulation (EU) No 2020/1197 of 30 July 2020. The Regulation sets the framework for the collection of R&D statistics and specifies the main variables of interest and their breakdowns at predefined level of detail.

31 October 2025

Please see the sub-concepts 3.4.1 and 3.4.2 in the full metadata view.

The statistical unit for BERD is the enterprise as defined by Council Regulation (EEC) No 1993/696 of 15 March 1993, if there are deviations please explain.

The statistical unit is the enterprise but the reporting units are legal units.

Please see the sub-concepts 3.6.1 and 3.6.2 in the full metadata view.

Not requested. R&D statistics cover national and regional data.

Calendar year 2023.

Accuracy in the statistical sense denotes the closeness of computations or estimates to the exact or true values. Statistics are not equal with the true values because of variability (the statistics change from implementation to implementation of the survey due to random effects) and bias (the average of the possible values of the statistics from implementation to implementation is not equal to the true value due to systematic effects).

Several types of statistical errors occur during the survey process. The following typology of errors has been adopted:

1. Sampling errors. These only affect sample surveys. They are due to the fact that only a subset of the population, usually randomly selected, is enumerated.
2. Non-sampling errors. Non-sampling errors affect sample surveys and complete enumerations alike and comprise:

• • Coverage errors,
  • Measurement errors,
  • Non response errors and
  • Processing errors.

Model assumption errors should be treated under the heading of the respective error they are trying to reduce.

R&D expenditure is published in the following units:  Euro (MIO_EUR) and Euro per inhabitant (EUR_HAB); data are available in the following units: basic unit National currency (MIO_NAC); Purchasing Power Standard (MIO_PPS); Purchasing Power Standard at 2005 prices (MIO_PPS_KP05); Purchasing Power Standard per inhabitant at constant 2005 prices (PPS_HAB_KP05); Percentage of gross domestic product (PC_GDP); and Percentage of total R&D expenditure (PC_TOT - for the breakdown by source of funds).

R&D personnel data are published in full-time equivalent (FTE), in head count (HC), as a percentage of total employment and as a percentage of active population.

See below.

Several separate activities are used for the collection of raw data or pre-compiled administrative data and statistics related to R&D. This section collects information on the type of data collection instruments used as well as methodological information for each data collection instrument. Depending on the type of data collection instrument used, only the sections corresponding to that data collection instrument are filled in.

Results based on the R&D survey are published biennially, in uneven years.
In even years Flanders publishes a 3% note with core R&D numbers (R&D expenditure, R&D personnel) on June 30 based on responses obtained in the innovation survey (CIS).

Timeliness and punctuality refer to time and dates, but in a different manner: the timeliness of statistics reflects the length of time between their availability and the event or phenomenon they describe. Punctuality refers to the time lag between the release date of the data and the target date on which they should have been delivered, with reference to dates announced in the official release calendar.

This sub-concept refers to the geographical comparability of data among the 27 Member States and the EFTA and Candidate Countries.

For more information related to the break years and the nature of the breaks, see the following sub-concepts in the full metadata view.