Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
Structural business statistics (SBS) describes the structure, conduct and performance of economic activities, down to the most detailed activity level (several hundred economic sectors). SBS covers all activities of the business economy with the exception of Agriculture, forestry and fishing activities, Public administration and non-market services. Main characteristics (variables) of the SBS data category include:
"Business demographic" variables (e.g. Number of active enterprises)
"Output related" variables (e.g. Net turnover, Value added)
"Input related" variables: labour input (e.g. Number of employees and self-employed persons, Hours worked by employees); goods and services input (e.g. Purchases of goods and services); capital input (e.g. Gross investments)
Business services statistics (BS) collection contains harmonised statistics on business services. From 2008 onwards BS become part of the regular mandatory annual data collection of SBS. The BS’s data requirement includes variable “Turnover” broken down by products and by type of residence of client.
The annual regional statistics collection includes three characteristics due by NUTS-2 country region and detailed on NACE Rev 2division level (2-digits).
3.2. Classification system
Statistical Classification of Economic Activities in the European Community (NACE): NACE Rev.2 is used from 2008 onwards. Key data were double reported in NACE Rev.1.1 and NACE Rev.2 only for 2008. From 2002 to 2007 NACE Rev. 1.1 was used and until 2001 NACE Rev.1
Starting reference year 2021 onwards, the SBS cover the economic activities of market producers within the NACE Rev. 2 Sections B to N, P to R and Divisions S95 and S96. Until 2007 the SBS coverage was limited to Sections C to K of NACE Rev.1.1 and from the reference year 2008 to 2020 data was available for Sections B to N and Division S95 of NACE Rev.2. From 2013, as the first reference year, to 2020 information is published on NACE codes K6411, K6419 and K65 and its breakdown.
From 2008 reference year data collection for the 'Business Services' covers NACE Rev 2 codes: J62, N78, J582, J631, M731, M691, M692, M702, M712, M732, M7111, and M7112.
3.4. Statistical concepts and definitions
SBS constitutes an important and integrated part of the new European Business Statistics Regulation N° 2152/2019
Surveying all legal units belonging to a complex enterprise
no
Surveying all legal units within the scope of SBS belonging to a complex enterprise
no
Surveying only representative units belonging to the complex enterprise
no
Other criteria used, please specify
Despite having no strict rule on surveying complex enterprises, all legal units with 20 or more employment (employees and self-employed) are always surveyed. This also applies to any legal units that generated a revenue value in the preceding year exceeding the 30th cumulative revenue percentile of the non-financial business population.
Comment
The sample frame is based on the legal structure of businesses in Malta. Consolidation into Enterprises is carried out towards the end of the SBS process. This includes integrating legal units that were not included in the original sample but that are consolidated within an enterprise together with sampled units, taking into account the implicit results at the legal unit level.
3.5.2. Consolidation
Consolidation method
Consolidation carried out by the NSI
yes
Consolidation carried out by responding enterprise/legal unit(s)
no
Other methods, please specify
Not available
Comment
The SBS is first compiled and scaled through grossing up at the legal unit level before being consolidated at the Enterprise Statistical Unit. Automatic profiling is applied to Enterprise Groups with 3 or fewer distinct activities. Enterprise Groups owning any legal unit that generates a turnover value above the 60th percentile of the total population turnover are subjected to manual profiling, irrespective of the number of legal units they own. Units not meeting the criteria for automatic profiling, such as Enterprise Groups with 4 or more activities, are also manually profiled. Through profiling, legal units are consolidated to form enterprises, with intra-enterprise transactions netted out, as these represent non-market transactions.
3.6. Statistical population
SBS cover selected market activities; more specifically NACE Rev. 2 Sections B to N, P to R, S95 and S96. The data includes local branches of foreign enterprises but does not include overseas activities controlled by Maltese enterprises. Essentially, the SBS seeks to measure the domestic formal economy. The target population is selected from the Statistical Business Register maintained by the NSO.
Starting from reference year 2021, the cut-off threshold of €7,000 in turnover, previously applied to the target population in past SBS datasets, was permanently eliminated.
3.7. Reference area
SBS data covers active resident enterprises in Malta and Gozo and Comino. This data is classified as a single region at the NUTS2 level, making the data at NUTS1 and NUTS2 level equivalent.
3.8. Coverage - Time
2008 - 2023.
3.9. Base period
Not applicable.
Number of enterprises and number of local units are expressed in units.
Monetary data are expressed in millions of €.
Employment variables are expressed in units.
Per head values are expressed in thousands of € per head.
Ratios are expressed in percentages.
2023
An alignment with the calendar year is typically observed for data collection and reporting. However, when the financial statements of an Enterprise do not follow a December year-end, the reporting is adapted to correspond with the fiscal year as specified in the respective audited set of financial statements. This policy is established to ensure the integrity of data, reflecting the actual financial period and to avoid subjective calendar adjustments of the financial data.
6.1. Institutional Mandate - legal acts and other agreements
Starting with reference year 2021 two new regulations currently form the legal basis of SBS:
Regulation (EU) 2019/2152 repealing 10 legal acts in the field of business statistics (EBS Regulation), and
The Council Regulation No 58/97 has been amended three times: by Council Regulation No 410/98, Commission Regulation No 1614/2002 and European Parliament and Council Regulation No 2056/2002. As a new amendment of the basic Regulation it was decided to recast the Regulation No 58/97 in order to obtain a new "clean" legal text.
In 2008 the European Parliament and Council adopted Regulation No 295/2008 and the provisions of this Regulation were applicable from the reference year 2008 to reference year 2020. Regulation No 295/2008 was amended by Commission Regulation (EU) No 446/2014.
6.2. Institutional Mandate - data sharing
Not applicable.
7.1. Confidentiality - policy
The NSO requests information for the compilation of official statistics according to the articles of the MSA Act – Cap. 422 and the Data Protection Act – Cap. 586 of the Laws of Malta implementing the General Data Protection Regulations (GDPR).
Article 40 of the MSA Act stipulates the restrictions on the use of information while Article 41 stipulates the prohibition of disclosure of information. Furthermore, Section IX of the Act (Offences and Penalties) lays down the measures to be taken in case of unlawful exercise of any officer of statistics regarding confidentiality of data.
Since its inception, the NSO has been committed to ensure the confidentiality of all collected data, using it solely for statistical purposes in accordance with established laws. The identity of data providers is protected, and any data that could lead to the identification of individuals or entities is not shared with third parties.
All NSO employees, upon joining, are made aware of confidentiality rules and requirements. As stipulated by the MSA Act, each staff member takes an oath of secrecy before starting their work.
The NSO has an internal policy concerning anonymisation and pseudo-anonymisation. This policy ensures that the data the office collects and shares is protected. It offers guidelines to NSO staff on applying appropriate data protection methods. This policy covers all data, irrespective of its format, managed by the NSO.
7.2. Confidentiality - data treatment
Primary confidentiality is applied in two situations:
When there are fewer than three enterprises.
When data from one business unit can approximate a component of another unit within 10% accuracy (the p% rule).
Secondary confidentiality protects primary confidential data. Data flagged as primary confidential are suppressed to avoid revealing sensitive information. The NSO identifies this data using a method consistent with other statistical agencies.
The confidentiality for tabular data is automated through two modular scripts. The primary script assigns confidentiality flags based on the threshold rule and the p-percent rule. The secondary script iteratively assigns additional flags using validation rules derived from classification hierarchies in the table, to prevent disclosure through deduction.
Starting from reference year 2021, the NSO introduced a new measure to reduce the use of confidentiality flags. Data from companies whose core metrics are disclosed in publicly available audited financial statements are deemed public and are not flagged for confidentiality, since such information is already in the public domain.
In addition, variables related to the number of employees and the number of enterprises are deemed non-confidential, on the basis that these metrics are already disseminated through other public sources.
7.2.1. Confidentiality processing
Data treatment
Confidentiality rules applied
yes. The p% rule is used
Threshold of number of enterprises (Number)
2
Number of enterprises non confidential, if number of employments is confidential
no
Dominance criteria applied
no
If dominance criteria applied specify the threshold (Number)
not applicable
Secondary confidentiality applied
yes
Comment
In cases where there should be a confidentiality flag but the variables are made public through the company's audited financial statements, confidentiality is not applied.
8.1. Release calendar
The NSO publishes and disseminates news releases at 11:00 hrs as scheduled in the Advance Release Calendar. The calendar is published on the NSO website and includes a three-month advance notice (the current month and the forthcoming two months). This calendar is sometimes subject to changes.
An internal policy on dissemination is in place to govern the dissemination of official statistics in an impartial, independent and timely manner, making them available simultaneously to all users.
The NSO’s primary channel for the dissemination of official statistics is the NSO website. Tailored requests for statistical information may also be submitted through the NSO website.
The results emanating from the SBS are used by other statistical domains and are reflected in other related news releases and publications.
10.2. Dissemination format - Publications
Not applicable.
10.3. Dissemination format - online database
Not applicable.
10.4. Dissemination format - microdata access
Not applicable.
10.5. Dissemination format - other
The data is transmitted to Eurostat, either to be used in European aggregates or to be used for ad-hoc analyses and/or satellite accounts.
10.6. Documentation on methodology
Work processes and procedures for the compilation of Structural Business Statistics are documented in a standardised reporting template and aligned to the GSBPM model. The model covers all phases of the statistical production process, from the initial stages of identifying what statistics are needed and the scope of the particular survey, to the final stages of dissemination and evaluation. The GSBPM report is only available internally and may be accessed by all NSO employees.
The Recommendations Manual on SBS provided important guidelines on the implementation of EBS into reference year 2021. EC Regulations No. 250 and 251/2009 also provide a set of guidelines for compiling SBS data.
The Structural Business Statistics Unit of Malta annually archives detailed records of its processes and quality control procedures for internal reference.
In alignment with the ESS Code of Practice (ESS CoP) standards, the NSO established a comprehensive Internal Quality Management Framework (QMF). This framework is supported by a document detailing general quality guidelines that apply across all statistical domains. While ensuring methodological soundness remains a central focus of the QMF, the document also addresses various institutional aspects.
11.1. Quality assurance
As part of the quality assurance procedure, the results of the SBS at macro level are compared with data from previous year. Any significant fluctuations in primary variables are closely examined, and the reasons for these variations are documented. When feasible, data is cross-verified with administrative sources to confirm their accuracy. The NSO consistently prioritises the accuracy of the data released to the public.
Every five to seven years, the NSO undergoes a Peer Review from Eurostat officials. This review gauges the NSO's alignment with the principles of the ESS CoP. These Peer Reviews are integral to the European Statistical System (ESS) strategy, enforcing adherence to the ESS CoP. Each National Statistical Institute (NSI) is prompted to complete a standard self-assessment questionnaire and this is followed by a visit of an expert team at the NSI, meeting its representatives and key stakeholders. The review culminates in a compliance report and highlights Improvement Actions, which the NSI is expected to act upon. The most recent Peer Review took place in 2022.
11.2. Quality management - assessment
The quality management process starts with vetting the online submissions of the respondents. These are checked thoroughly to identify item non-response which is imputed from administrative sources, if available and also for other logical errors. The respondents may be contacted and queried about particular responses. After the data collection process, the data goes through a stringent quality framework in four different phases.
Phase 1 - A number of sector specific ratios (reflecting those applied at the Conval stage in the Input Hall) are applied on the micro data. Units that fall outside the acceptable thresholds need to be confirmed or updated as necessary. Significant growths in the main variables when compared to the previous year are also scrutinised. New units that were not in the sample of the previous year and units which were in the sample of the previous year but not of the current are also examined.
Phase 2 - The main variables are checked against those computed from administrative data sources to check for any possible errors in the data collected for the individual units. The trade margins and mark ups of units in the wholesale and retail trade are checked against the average margins over a time-series from 2008 for their particular NACE category. Those which vary significantly from the average are checked and confirmed or updated as necessary. Units with high values of GVA/head and wage/head are investigated to ensure that the data at micro level is accurate.
Phase 3 -To ensure that the grossed up dataset is accurately representing the larger population, specifically units for which data was not collected, the implicit values for these units are identified using the methodology of the grossing up process. These implicit values are subsequently compared with values from administrative data sources. If significant discrepancies are detected between the two, the unit in question is subjected to imputation and is included as an observation unit but remains excluded from the sampling pool since such units are normally anomalies to the average performance of the respective stratum. In this manner, such additions would only represent themselves and would not influence the overarching patterns of the respective stratum.
Phase 4 - The GVA of industries at NACE Division level is compared to the values of the previous year and the main observations are documented. Reasons for growths or declines are established with special attention being given to main contributors to such events.
12.1. Relevance - User Needs
External users of SBS data include Eurostat, the European Commission, researchers, students, international and local organisations focusing on SMEs. Internal users include other units within the NSO mainly the National Accounts, Balance of Payments, Environment, Agriculture and Fisheries and the Business Registers, Research and Innovation units .
12.2. Relevance - User Satisfaction
The last User Satisfaction Survey was held in 2014 with the aim to collect information about key users’ satisfaction with statistical output.
The NSO keeps record of the number of News Releases and publications disseminated on its website; the users to whom statistical products are provided; as well as the number of requests that are processed every year.
News Releases and tailor-made statistical outputs were assessed on account of their quality, timeliness, and on their ability to meet users’ needs.
12.3. Completeness
All data requested in the SBS Regulation and which is relevant to Malta are regularly provided. The completeness assessment is rated in the EU-wide Quality Report produced by Eurostat.
Malta is usually exempt from sending variables that are less than the one percent of the European Community total and thus makes use of the 1% rule (exemption) and the CETO flag.
13.1. Accuracy - overall
The main sources of error are assumed to be linked with the non-response rate, a controlled level of sampling error, and the respondents’ understanding of the questionnaire (non-sampling errors), with the biggest potential errors being non-sampling errors.
13.2. Sampling error
The SBS methodology in Malta is fundamentally rooted in the ratio estimator approach to calculate Gross Value Added (GVA) and other key SBS variables. This method identifies performance patterns in the sampled data, such as Intermediate Input productivity, and applies these patterns by stratification level to estimate GVA across the unsampled portion of the population. This essentially measures how efficiently turnover is converted into value added, by stratum observations, and is used as a benchmark to estimate missing data. Due to this methodology, as opposed to the traditional weight-assigning method, the accuracy of the SBS results of Malta is not best captured by traditional parameters like the Coefficient of Variation. Instead, accuracy is assessed using the margin of error (at 95% confidence level) based on the standard deviation of Intermediate Input productivity to evaluate the GVA's accuracy relative to the total population turnover. For 2023, the total non-financial economy sampling margin of error for the unsampled portion of the population was 3.97% of GVA (18 months after the end of the reference period).
From reference year 2023, Malta applies a revised methodology for the calculation of the coefficient of variation (CoV) in SBS. The change aims to align the indicator more closely with the structure of the actual SBS grossing-up process and to provide users with a clearer measure of sampling uncertainty. The revised CoV is therefore applied only to the relative proportion of the result that is subject to sampling and grossing-up, rather than providing an overall variance snapshot of the industry structure, as it did in the past.
The method is implemented directly on the SBS grossing-up strata. Census-based variables (Enterprises, Turnover, Employment) are assigned a CoV of zero, as no grossing-up is involved. For GVA, which is partly derived through sampling, the CoV is based on the turnover-to-GVA efficiency rate observed in the sample and applied through the ratio estimator. Confidence intervals are derived at stratum level and aggregated upwards to the levels required for EU regulation-specific groupings.
This approach is aimed for the published CoV values to be consistent with Malta’s SBS methodology and to provide users with a transparent measure of data quality by indicating the extent of potential sampling error.
13.3. Non-sampling error
The sources giving rise to non-sampling error include the following:
Data providers providing data for a different reference year than the one required;
Data not converted to Euro;
Improper record-keeping of important variables (such as hours worked);
Number of employees not provided as an average for the year;
Incorrect values provided for intra-company transactions; and
NACE misclassifications.
In practice, these errors are relatively rare and generally have limited impact. The main source of non-sampling error arises from imputed estimates for large enterprises. While most imputations are based on administrative sources, some rely on modelled estimates. When no administrative data describing the performance of an enterprise are available at the time, the unit’s activity and size class are taken into account, and an intelligent hot-deck imputation method is applied. For reference year 2023, estimated imputations in the non-financial business economy accounted for 4.8% of total GVA and 6.1% of the GVA of observed units.
The assessment on potential inaccurate reporting by respondents and/or businesses and wrong interpretation of financial statements cannot be accurately measured or estimated.
14.1. Timeliness
Preliminary SBS data, transmitted to Eurostat 10 months after the end of the reference period, is not published on the NSO (Malta) website due to the very limited information available at that stage. SBS data is transmitted to Eurostat at T+18 months and a summary of the full SBS results is published after the full transmission to Eurostat.
14.2. Punctuality
SBS 2023 was provided to Eurostat on 01 July 2025.
15.1. Comparability - geographical
The data is comparable across other EU countries due to common definitions and methodology of data.
15.2. Comparability - over time
The definitions and regulation align for years 2008-2020. Reference years from 2021 onwards are based on EBS, whereby a broader coverage of industries is covered as well as other methodological changes including the calculation of certain variables and the transmission methods. In addition to the EBS implementation, SBS data is now compiled on the basis of the Enterprise Statistical Unit.
15.2.1. Time series
Time series
First reference year available (calendar year)
2008
Calendar year(s) of break in time series
2021
Reason(s) for the break(s)
Implementation of EBS
Removal of €7,000 turnover threshold on the target population
The introduction of the Statistical Unit Enterprise
Length of comparable time series (from calendar year to calendar year)
2008–2020, 2021–2023
Comment
15.3. Coherence - cross domain
In general, SBS data follows the same principles as Short-term business statistics, Statistical business registers, and National accounts. Differences in outputs can occur mainly due to diverging objectives, timeliness and range of possible inputs.
15.4. Coherence - internal
Computed fields were consistent across all years, but with the implementation of the EBS regulation, some changes in the computation of the variables created a break-in series in the comparability of these variables. Such an example is the computation of Value added.
SBS data seeks to be coherent with the Business Demography data also in fulfilment of the EBS regulation obligation. Internally, coherence is sought between SBS and National Accounts data.
For reference year 2023, around 8,300 enterprises made up the sample of SBS. Response was required from 5,167 enterprises as the rest were retrieved entirely from administrative sources. The SBS Unit constantly seeks ways to reduce administrative burden primarily by shortening the questionnaire for specific groups and replacing the answers of several questions using available administrative information. Self-employed business units whose data is available in administrative data sources, are required to fill in a shorter version of the questionnaire, which only expects the response for questions related to employment and investments. Additionally, companies are given the opportunity to upload a full set of financial statements and only fill in a limited number of questions.
A group of around 3,100 companies were not contacted, since all their data could be taken from financial statements which are publicly available and other administrative data sources.
17.1. Data revision - policy
At the NSO, the revisions policy may be accessed from this link.
At the SBS Unit, the revision policy set for the SBS domain is set to revise all data one year after the target date. This review at T+30 (months) concept is a voluntary action programmed as a measure of quality for boosting the accuracy of SBS results to Eurostat and concurrently enhance the coherence between the National Accounts Statistics and the Structural Business Statistics.
17.2. Data revision - practice
SBS data is scheduled for review at T+30 months and transmitted to Eurostat. After such review exercise, data is considered final.
18.1. Source data
SBS data is collected through a statistical survey, which is heavily supplemented and cross-checked with various administrative data sources and observations from other NSO departments. The Neyman Allocation is the chosen sampling strategy, which determines the required sampling counts for each stratum based on both the volume of units in the sampling frame and the efficiency variance observed in the preceding year's results. Strata are defined by employment size classes and NACE Divisions, though in certain instances, categorisation extends down to NACE Groups or Classes. The sampling frame, from which the sample is selected, and the target population are not identical. This is because they are extracted from the Statistical Business Register at different times and the sampling frame specifically excludes units with a turnover below €10,000 due to a predetermined threshold.
18.1.1. Data sources overview
Data sources overview
Survey data
yes
VAT data
yes
Tax data
yes
Financial statements
yes
Other sources, please specify
Other units within the NSO
Comment
Not applicable
18.2. Frequency of data collection
Annual data collection.
18.3. Data collection
The SBS questionnaire is delivered to statistical units electronically. Initially, a letter is dispatched to the business address to confirm the email addresses. A few days later, unique links and passwords to access the e-questionnaire are sent via email. For those business units without internet access, a paper questionnaire is mailed. A .pdf version of this questionnaire is also available on the NSO website.
Non-respondents receive reminders through both emails and letters, followed by telephone calls for further follow-up. Data pertaining to credit institutions is sourced from the Central Bank of Malta and other administrative data sources to be compiled and transmitted by the NSO to Eurostat.
18.4. Data validation
During the vetting phase, data is checked for obvious errors or item non-response and cross-referenced with financial statements. In addition, the information provided is analysed against data from previous years to identify any significant fluctuations. After the data is collected and vetted, it passes through a number of quality control checks to ensure that the data at micro level is within the set thresholds and that growth rates are confirmed.
The data verification rules include the SBS validation rules incorporated in Eurostat’s validation tool (InputHall). The reasons for the remaining deviations are identified and described in a report submitted to Eurostat during the data validation process.
18.5. Data compilation
In case of non-response, administrative sources are used to replace the questionnaire information. When administrative data is unavailable, cold-deck imputations are primarily used, while also factoring-in the relevant stratum performance patterns from the current reference year. The grossing-up mechanism is deployed at stratum level. Stata are allocated based on the NACE Rev.2 framework (primarily at the two-digit level), and employment size categories. Incoming data is cross-referenced with Financial Statements (if accessible) and any available administrative sources. Any inquiries or discrepancies are addressed directly with respondents, typically via email or phone.
18.6. Adjustment
The NSO adjusts responses of hours worked that do not make reasonable sense and are above a certain threshold. For example, some companies mechanically report 2,080 hours per employee (52 weeks × 40 hours), without accounting for public holidays, vacation or sick leave. Other cases involve values exceeding 2,100 hours per full-time equivalent employee, which would imply that employees worked more than 40 hours every week of the year, again without any leave. In principle such responses may indicate that the respective businesses do not have an accurate measurement tool for actual hours worked. These responses are updated with more reasonable values that reflect the more likely actual hours spent at the place of work. Such adjustments are based on the results and indicators of the labour force survey (LFS) for the reference year, at NACE Section level.
No further comments
Structural business statistics (SBS) describes the structure, conduct and performance of economic activities, down to the most detailed activity level (several hundred economic sectors). SBS covers all activities of the business economy with the exception of Agriculture, forestry and fishing activities, Public administration and non-market services. Main characteristics (variables) of the SBS data category include:
"Business demographic" variables (e.g. Number of active enterprises)
"Output related" variables (e.g. Net turnover, Value added)
"Input related" variables: labour input (e.g. Number of employees and self-employed persons, Hours worked by employees); goods and services input (e.g. Purchases of goods and services); capital input (e.g. Gross investments)
Business services statistics (BS) collection contains harmonised statistics on business services. From 2008 onwards BS become part of the regular mandatory annual data collection of SBS. The BS’s data requirement includes variable “Turnover” broken down by products and by type of residence of client.
The annual regional statistics collection includes three characteristics due by NUTS-2 country region and detailed on NACE Rev 2division level (2-digits).
29 August 2025
SBS constitutes an important and integrated part of the new European Business Statistics Regulation N° 2152/2019
SBS cover selected market activities; more specifically NACE Rev. 2 Sections B to N, P to R, S95 and S96. The data includes local branches of foreign enterprises but does not include overseas activities controlled by Maltese enterprises. Essentially, the SBS seeks to measure the domestic formal economy. The target population is selected from the Statistical Business Register maintained by the NSO.
Starting from reference year 2021, the cut-off threshold of €7,000 in turnover, previously applied to the target population in past SBS datasets, was permanently eliminated.
SBS data covers active resident enterprises in Malta and Gozo and Comino. This data is classified as a single region at the NUTS2 level, making the data at NUTS1 and NUTS2 level equivalent.
2023
An alignment with the calendar year is typically observed for data collection and reporting. However, when the financial statements of an Enterprise do not follow a December year-end, the reporting is adapted to correspond with the fiscal year as specified in the respective audited set of financial statements. This policy is established to ensure the integrity of data, reflecting the actual financial period and to avoid subjective calendar adjustments of the financial data.
The main sources of error are assumed to be linked with the non-response rate, a controlled level of sampling error, and the respondents’ understanding of the questionnaire (non-sampling errors), with the biggest potential errors being non-sampling errors.
Number of enterprises and number of local units are expressed in units.
Monetary data are expressed in millions of €.
Employment variables are expressed in units.
Per head values are expressed in thousands of € per head.
Ratios are expressed in percentages.
In case of non-response, administrative sources are used to replace the questionnaire information. When administrative data is unavailable, cold-deck imputations are primarily used, while also factoring-in the relevant stratum performance patterns from the current reference year. The grossing-up mechanism is deployed at stratum level. Stata are allocated based on the NACE Rev.2 framework (primarily at the two-digit level), and employment size categories. Incoming data is cross-referenced with Financial Statements (if accessible) and any available administrative sources. Any inquiries or discrepancies are addressed directly with respondents, typically via email or phone.
SBS data is collected through a statistical survey, which is heavily supplemented and cross-checked with various administrative data sources and observations from other NSO departments. The Neyman Allocation is the chosen sampling strategy, which determines the required sampling counts for each stratum based on both the volume of units in the sampling frame and the efficiency variance observed in the preceding year's results. Strata are defined by employment size classes and NACE Divisions, though in certain instances, categorisation extends down to NACE Groups or Classes. The sampling frame, from which the sample is selected, and the target population are not identical. This is because they are extracted from the Statistical Business Register at different times and the sampling frame specifically excludes units with a turnover below €10,000 due to a predetermined threshold.
Annual.
Preliminary SBS data, transmitted to Eurostat 10 months after the end of the reference period, is not published on the NSO (Malta) website due to the very limited information available at that stage. SBS data is transmitted to Eurostat at T+18 months and a summary of the full SBS results is published after the full transmission to Eurostat.
The data is comparable across other EU countries due to common definitions and methodology of data.
The definitions and regulation align for years 2008-2020. Reference years from 2021 onwards are based on EBS, whereby a broader coverage of industries is covered as well as other methodological changes including the calculation of certain variables and the transmission methods. In addition to the EBS implementation, SBS data is now compiled on the basis of the Enterprise Statistical Unit.
