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.
ISTAT (Istituto Nazionale di Statistica) - National Statistical Institute of Italy
1.2. Contact organisation unit
Business Statistics Directorate /Division for Statistics on Production and Foreign Trade
1.3. Contact name
Restricted from publication
1.4. Contact person function
Restricted from publication
1.5. Contact mail address
ISTAT Via Tuscolana 1788 00173 Roma Italia
1.6. Contact email address
Restricted from publication
1.7. Contact phone number
Restricted from publication
1.8. Contact fax number
Restricted from publication
2.1. Metadata last certified
30 August 2024
2.2. Metadata last posted
30 August 2024
2.3. Metadata last update
30 August 2024
3.1. Data description
For reference year 2021 and onwards, the Commission Implementing Regulation EU 2020/1197 is laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council. According to the requirements in the EBS General Implementing Act, the following three variables for industrial production are required:
• Variable 251001 (Sold production): national currency and (except for industrial services) quantity as defined in the PRODCOM List in force at the end of the reference period.
•Variable 251002 (Production under sub-contracted operations): (except for industrial services) national currency and quantity as defined in the PRODCOM List in force at the end of the reference period.
• Variable 251003 (Actual production): quantity as defined in the PRODCOM List in force at the end of the reference period. The survey sample for reference year 2023 is little less than 42,000 enterprises with approximately 65,000 local units, covering about 230 industries, derived from sub-sectors of the total manufacturing sector, and about 4,000 products (that include also industrial services). With the entry into force of the EBS (EU) regulation on European business statistics, the inclusion of the KAU (Kind of activity unit) statistical unit in the Registers managed by the National Statistical Institutes becomes mandatory. Consequently, this unit of analysis is also adopted in the PRODCOM survey.
3.2. Classification system
In PRODCOM classification, products are grouped by the NACE Rev.2, 4-digits industrial sectors, in which they are primarily produced (i.e. characteristic products). The value and volume data shown for a product, therefore, includes not only the production by enterprises classified to the NACE sector where it is primarily produced but also its secondary production by enterprises classified to the other NACE Rev.2 sectors. Classification of Products: The PRODCOM list is divided into "Classes" corresponding to the classes of NACE Rev.2. Within the individual classes, the headings of the list are set out in ascending order of their PRODCOM codes. PRODCOM uses an eight-digit product code of the form XX.XX.XX.YY. The first four digits of the code correspond to the 4-digit classes of NACE Rev.2. The first six digits are the Community Classification of Products by Activity (CPA) codes. The CPA provides a detailed listing of the characteristic products for each 4-digit NACE Rev.2 economic activity. The last two digits provide a more detailed breakdown of the CPA classes into PRODCOM product headings. There is a direct link between the PRODCOM classification and the EU foreign trade Combined Nomenclature (CN). Each PRODCOM 8-digit heading covers one or more CN 8-digit headings. At national level: To view the national classification, please see ATECO2007.
3.3. Coverage - sector
The annual survey covers all industrial enterprises with 3 or more employees, engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries. Under PRODCOM, the NACE Rev.2 classification is having covered the sections B and C, comprising divisions 07-33, but excepting 09 and 19 and from reference year 2019, is included part of division 38. In order to fulfil the regulation provisions (to reach 90% representativeness), a probabilistic sample based survey strategy has been designed. It involves a sample of small industrial enterprises (from 3 to 19 employees) and the totality of units that belong to medium and large companies (20 employees and over). The sampling method is based on an equiprobabilistic selection in each stratum, obtained by the combination of the three variables: class of employees, class of economic activity and geographical breakdown. Variables that are used to evaluate the sample size in each stratum are: turnover, added-value and number of employees. The size is obtained by applying the Bethel method.
3.4. Statistical concepts and definitions
From 2021 onwards, the following information is requested for each variable:
The sold production (including industrial services) defined as sold (invoiced) production, which may be carried out under the primary or secondary activities of the enterprise. It includes production sold (invoiced) between different kind-of-activity units belonging to the same enterprise. The production under sub-contracted operations, which has been sold (invoiced) to the principal. The production may be carried out under the primary or secondary activities of the enterprise. The actual production includes any production carried out during the reference period and within the economic territory. With regard to subcontracting relationships we do not have the information of who is the owner of the goods.
3.5. Statistical unit
Generally, the reporting unit is the Legal Unit as a proxy for the enterprise; the observation unit is Local Unit(s). From 2021 onwards the reporting unit is the KAU. However, we are using profiling in order to follow some dominant enterprises.
3.6. Statistical population
The survey population is made up of all industrial enterprises in NACE 07-33 (excepting 09 and 19) and 3832, with more than 20 employees and a sample of small industrial enterprises (from 3 to 19 employees).The population is selected every year from Statistical Business Register.
3.7. Reference area
Italy - whole national territory.
3.8. Coverage - Time
The period observed is one calendar year. Italian industrial products statistics date back to 1996. For 1996 to 2007, the files contain the original data based on NACE Rev.1.1. For 2008 onwards, the files contain data that has been converted where possible to NACE Rev.2. Data are available on Eurostat’s website; new data Browser accessible at the link.
3.9. Base period
Not applicable.
For Volume: Physical measurement unit required by Prodcom list, specific for each product.
For Value: Thousand Euro (excluding VAT and excise duties).
Year 2023
6.1. Institutional Mandate - legal acts and other agreements
The survey is carried out in compliance with following regulations:
At European level:
EUR-Lex - 32020R1197 - EN - EUR-Lex (europa.eu) Commission Implementing Regulation (EU) 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.
EUR-Lex - 32019R2152 - EN - EUR-Lex (europa.eu) Regulation (EU) 2019/2152 of the European Parliament and of the Council of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics.
EU implementing regulation 2022/2552 on technical specifications of data requirements for the detailed topic industrial production statistics establishing the industrial products classification breakdown.
At national level:
Presidential Decree (DPR) of July 11, 2023, approving National Statistical Program (PSN) 2020-2022, updating 2022. The statistical survey is mandatory to be filled-in by the private units. More info about national law.
6.2. Institutional Mandate - data sharing
Not applicable.
7.1. Confidentiality - policy
According to the article n. 9 of the Legislative Decree n. 322 of 6 September 1989 data collected by statistical offices within the statistical surveys included in the National Statistical Programme may not be disclosed other than in aggregated form such that no reference to identifiable people can be extracted. Furthermore, they may be used only for statistical purposes. Data may not be communicated or disseminated neither to any external subject, public or private, nor to any department of the public administration other than in aggregate form and using modalities which prevent the identification of the people involved. In any case, data cannot be used to identify again the people involved. The Code of Conduct annexed to the Legislative Decree no. 196 of 30 June 2003 (Personal Data Protection Code) and Legislative Decree no. 101 of 19 September 2018 (General Data Protection Regulation) provides special rules concerning the processing of personal data for statistical purposes within Sistan. In order to make statistical secrecy confidentiality and protection of personal data effective, Istat is currently taking appropriate organizational, logistical, methodological and statistical measures in accordance with internationally established standards. In accordance with the Legislative Decree no. 196 of 30 June 2003 (Personal Data Protection Code) and subsequent modifications and additions, respondents are informed of their rights and obligations with regard to the provision of information, and they are assured that the information they provide will be used for statistical purposes only. Special aggregation rules have been developed to ensure that indirect disclosure of individual data does not occur when aggregations of data are presented. For instance, access to individual data is restricted to staff who require the information in the performance of their duties. Provisions are in place to supervise analysts that require access to disaggregated data.
The current legal framework governing European statistics is Regulation (EC) No 223/2009 that was adopted by the European Parliament and the Council in 2009 and that was revised in 2015 to strengthen the overall governance of the European Statistical System (ESS) and, notably, its professional independence.
7.2. Confidentiality - data treatment
The Statistics Italy's confidentiality policy for PRODCOM data established that a certain PRODCOM code will be flagged where there are less than 3 reporting companies (for each published indicator). For 2023, about 35% of all results of the Italian PRODCOM data are confidential.
Data are disseminated at national and at European level on yearly basis.
10.1. Dissemination format - News release
A statistic report is expected by the end of the year.
10.2. Dissemination format - Publications
Statistica annuale della produzione industriale - Anno 1996 (2000), Info: Col.informazioni Statistica annuale della produzione industriale - Anno 1997 (2001), Info: Col.informazioni Statistica annuale della produzione industriale - Anno 1998 (2001), Info: Col.informazioni Statistica annuale della produzione industriale - Anno 1999 (2002), Info: Col.informazioni
As part of microdata release, Istat offers a research data center: ADELE Laboratory (Laboratory for Elementary Data Analysis), a "secure" site accessible by researchers and academics enabling them to conduct their own statistical analyses on microdata from the Institute's surveys in compliance with legislation concerning the confidentiality of personal data .
10.5. Dissemination format - other
Tables: Statistica annuale della produzione industriale. Anno 2004
Tables: Statistica annuale della produzione industriale. Anno 2005
Tables: Statistica annuale della produzione industriale. Anno 2006
Tables: Statistica annuale della produzione industriale. Anno 2007
Tables: Statistica annuale della produzione industriale. Anno 2008
Tables: Statistica annuale della produzione industriale. Anno 2009
Tables: Statistica annuale della produzione industriale. Anno 2010
Tables: Statistica annuale della produzione industriale. Anno 2011
No documentation on quality management are available on NSI's site.
11.1. Quality assurance
Since ‘90s Istat adopted a systematic approach to ensure quality in both statistical information and service to the community. For this purpose, the Institute has defined a quality policy providing itself with appropriate tools as well as management changes to carry it out.
Istat quality policy is coherent with the European framework developed by Eurostat, taking up its main principles and definitions. In 2005, the European Statistics Code of Practice (revised in 2011) established the principles to follow in order to ensure and strengthen both accountability and governance of the European Statistical System and the National Statistical Systems within it.
Essential points of Istat quality policy are:
Process quality: consisting in the production of accurate statistical information efficiently and effectively; Product quality: consisting in the dissemination of high-quality timely statistical data which are relevant for the users, also the potential ones; Documentation: consisting in the storage and availability of information necessary not only for a proper use of data but also to ensure transparency in all the production activities of statistical data; Respect for respondents: consisting in the reduction of response burden and in the respect of respondent's privacy; Strengthening of statistical literacy: consisting in promoting a proper use of statistical information in policy-making to better support decisions and policies; Users' orientation: consisting in making statistical information easily accessible and understandable and in satisfying user needs as much as possible. For more details.
11.2. Quality management - assessment
The Information System provides a number of arithmetic and logical controls, both mandatory and optional between the items in each form and between the forms in the whole annual report. There are a number of arithmetic and logical controls inside the PRODCOM form. The System calculates both the totals and the unit price. The PRODCOM questionnaire cannot be completed successfully before all errors inside the questionnaire, detected by the checking program, to be corrected.
12.1. Relevance - User Needs
The main users of PRODCOM data can be presented in different groups, namely:
EXTERNAL/ INTERNATIONAL (OUTSIDE NSI): Eurostat;
EXTERNAL/ NATIONAL (OUTSIDE NSI), like: Institutions at national level - governmental (Ministry of Economy; Ministry of Agriculture and Food, ENEA the Italian National Agency for New Technologies, Energy and Sustainable Economic Development); Institutions at national level - non-governmental (Trade associations; Researchers and students, Companies);
INTERNAL (INSIDE NSI): Environmental statistics, National accounts, STS, Agriculture statistics, Business Register.
12.2. Relevance - User Satisfaction
No user satisfaction survey has been done for PRODCOM.
12.3. Completeness
In Italy are inquired all codes of the Procom list.
Quality indicator: 100%.
In advance there are some codes that exist only for Italian dissemination (are typical Italian production i.e. in the food industry).
13.1. Accuracy - overall
For reference year 2023, the response rate is 56,9% of all enterprises surveyed.
The main causes of non-response are: misclassification of economic activity (activities outside of sections B and C, NACE Rev.2), inactive, merger, split off, unreachable unit.
After checking if the company has stopped or suspended its activity during the reference year, the non-responses were estimated using the previous year data and data from other sources. In case of partial responses, that concerns basically missing volume data, estimates for missing volumes are done by using the median unit price from the other respondent companies.
13.2. Sampling error
A probabilistic sample has been designed. It involves a sample of small industrial enterprises (from 3 to 19 employees) and the totality of the local units that belong to medium and large companies (20 employees and over). For the selection of the probability sample, a stratified random sampling plan was adopted, at one stage, with constant selection probability for all the units belonging to the same stratum. The unit selection layer was constructed by concatenating the modes of the following variables: - Nace class (4 digits) - 2 classes of employees (3-19, 20 and over) - 5 territorial divisions (North-West, North-East, Centre, South, Islands). The optimal sample size is obtained by applying the Bethel method. In order to ensure adequate coverage, productive local units belonging to companies having main activities outside the industrial sectors covered by PRODCOM are included in the survey as well. The required coverage levels for PRODCOM survey is achieved by means of a grossing up technique. The estimation technique currently adopted makes use of a calibration method based on the auxiliary variables number of enterprises and number of employees belonging to industrial companies included in the target population. The sampling CV (coefficients of variation) of each statistic (added-value and number of employees) is estimated by the ratio between square root of sampling variance estimate and parameter estimate in each estimation domain.
Coefficient of Variation = 0.05 for each statistic (added-value and number of employees) (for the formula see ESS handbook for quality and metadata reports Part C Guidelines for ESS Quality and Performance Indicators). This value is set in the phase of the sampling selection using the Bethel procedure.
13.3. Non-sampling error
Measurement errors: Respondents have problems with the questions about the produced volume, sometimes they have to provide estimations.
Non-response errors: After the first publication, new response will cause adjustments.
Response rates: Unweighted response rate (number of enterprises): 56,9%
Sold production value
Sold production quantity
Production under sub-contracted operations quantity
Production under sub-contracted operations value
Actual Production quantity
Unweighted imputation rate (number of items)
7,29
8,37
7,35
7,39
14,19
Item non-response: Item non-response rate has not calculated because would be heavily influenced by the accuracy of the selection made by Statistics Italian that provides for each respondent a selection of Prodcom product groups, which are expected to be relevant to the respondent, based on historical data and the NACE classification. This selection will often contain product groups in which the respondent has no production.The frame for unit Local units is Italian Statistical Business Register managed by Istat (National Statistical Institute of Italy).
The quality actions to control unit nonresponse are:
Survey presentation letter signed by Istat President Guarantees on statistical confidentiality Editing and imputation procedures are used to correct for item nonresponse and measurement errors Data are validated to the following actions: Coherence control with data from other surveys or sources Coherence control with previous data of the same survey Control of influential unit data, strongly contributing to computation of aggregates (e.g. large enterprises).
Undercoverage
There are target population units that are not accessible via the survey frame e.g. enterprises created, merged or split in the reference year of the survey, enterprises misclassified (wrong activity or missing secondary activities)
Overcoverage
There are units accessible via the survey frame that do not belong to the target
When an enterprise is found to be out of the scope of the survey, e.g. it has no productive activity, then the enterprise is no longer surveyed under the PRODCOM survey. The over-coverage rate for the survey referring to 2023 is 2,0%
Multiple listings
No units are present more than once in the frame. Each LKAU is characterized by an identity number relating to name and address of the unit.
14.1. Timeliness
Data are disseminated at national level Yearly.
Data are provided to Eurostat Yearly within t+6 months.
14.2. Punctuality
In Italy, the PRODCOM data are usually published without delay in relation to the scheduled date.
15.1. Comparability - geographical
Data are comparable with those of other European Union (EU) countries.
All EU countries produce PRODCOM statistics, which is the system for the collection and dissemination of statistics of industrial goods based on the PRODCOM List (8-digit codes).
In Italy, the List of Products used is based on the PRODCOM List with Italian codes added according to the interest of users.
15.2. Comparability - over time
With implementation of the EBS Regulation, a break of series due to the fact that the production subject to subcontracted operations was collected separately from own-account production from the year 2021.
From a NACE classification perspective, the introduction of NACE Rev. 2 in 2008 resulted in a break in series.
15.3. Coherence - cross domain
For each domain, the coherence with previous data is checked.
Cross-checks are made with Business Register, SBS and STS at micro and macro level. Cross-checks with ITGS are planned to be introduced.
Coherence with the Turnover Index in Industry is checked by comparing the percentage change of the sales value between the reference and the previous year at 2-digit level of NACE Rev. 2, with the respective change of the Turnover Index in Industry.
However these comparisons are just an indication for the trend of the production over time because of different frequencies and methodological standards.
15.4. Coherence - internal
PRODCOM statistics are internally coherent.
The survey is based on the Prodcom list which defines the products to be collected. Data on these products, obtained by producer enterprises, are then aggregated at the product level (8-digit code).
The online survey and validation checks ensure coherence between variables. Quantities, values of production and unit prices of product are also checked in time series using an interval of 5 years.
Average company's time to fill in the survey: about 2 hours.
Actions to reduce burden: pre-filled questionnaire and single sign (access with a unique user id and password), using enterprise portal.
17.1. Data revision - policy
The Italian revision policy lays down standard rules and principles for data revisions, in accordance with the European Statistics Code of Practice and the principles for a common revision policy for European Statistics contained in the Annex of the European Statistical System (ESS) guidelines on revision policy.
17.2. Data revision - practice
The data are sent to Eurostat by the end of June (T+6). First revision is planned in September (T+9). If a last revision is necessary, it is carried out in December (T+12).
For previous years, if errors are found, data are revised and provided.
18.1. Source data
The survey sample is selected from Statistical Business Register 2021. PRODCOM estimates cover Italy’s registered businesses for 90% (we made census for enterprises with 20 or more employees, which totals over 30.000 units).
For the enterprises with less than 20 employees we made a sample of over 10.000 units. VAT and turnover is classified to NACE divisions 07-33 and 38. Businesses are selected by employment size, industry classification and geographical breakdowns, using a stratified random sample. Additional constraints are built into the sample design so to decrease the form filling response burden on smaller businesses. The number to be selected has been calculated to minimize the burden of sampling, without compromising the calculation of final estimates.
The sampling method is based on an equiprobabilistic selection in each stratum obtained by the combination of two three variables: class of employees, class of economic activity and geographical repartitions. The variables used to evaluate the sample size in each stratum are: turnover and number of employees. The determination of sample size and its allocation between the stratum occurred requiring that the expected error in the estimated total (error expressed in terms of coefficient of variation) of the variables considered were less than 5 percent in each domain of study represented by the class of economic activity. The allocation of the sample in the layers was achieved by adopting a multivariate allocation algorithm proposed by Bethel, which is a generalization of the method of Neymann for only one variable, and that allows to minimize the sample size so as to satisfy the constraints precision of the estimates for each of the variables considered of interest, relative to the domain estimation.
18.2. Frequency of data collection
On a yearly basis.
18.3. Data collection
Self-administered data collection by e-questionnaire (Computer Assisted Web Interviewing-CAWI)
The web questionnaire and instructions for companies are available.
The quality actions to control unit nonresponse are:
Survey presentation letter signed by Istat President
Guarantees on statistical confidentiality
Coercive measure: the answer to the statistical process is mandatory with penalties for enterprises with over 250 employees.
18.4. Data validation
Data are validated to the following actions:
Coherence control with previous data of the same survey Control of influential unit data, strongly contributing to computation of aggregates (e.g. large enterprises).
18.5. Data compilation
Steps of data compilation in PRODCOM statistics of Italy:
1. Error detection based on consistency edits
2. Manual error correction forcing wrong values on the base of deterministic rules as reported in a manual or document
3. Manual error correction forcing wrong values on the base of previous correct data of the same survey
4. Manual error correction based on reporting unit follow-up
5. Manual coding assisted by paper lists (manuals, listings) or computer screens
6. Data entry by Istat section in charge of the survey
7. Splitting questionnaire according to various record types
8. Deterministic error and outlier detection and imputation based on deterministic rules (IF-THEN)
9. Error and outlier detection and imputation by clerical reviewers based on experience
10. Error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
11. Error and outlier detection and imputation by clerical reviewers based on questionnaire review
12. Error and outlier detection and imputation by clerical reviewers based on respondent follow-up
13. Deterministic error and outlier detection and imputation by clerical reviewers based on experience
14. Deterministic error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
15. Deterministic error and outlier detection and imputation by clerical reviewers based on data from other surveys or sources
16. Deterministic error and outlier detection and imputation by clerical reviewers based on questionnaire review
17. Deterministic error and outlier detection and imputation by clerical reviewers based on respondent follow-up
18. Deterministic error and outlier detection and imputation by clerical reviewers based on the available auxiliary variables
19. Estimate calculation
18.6. Adjustment
Not applicable for PRODCOM data.
No comment.
For reference year 2021 and onwards, the Commission Implementing Regulation EU 2020/1197 is laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council. According to the requirements in the EBS General Implementing Act, the following three variables for industrial production are required:
• Variable 251001 (Sold production): national currency and (except for industrial services) quantity as defined in the PRODCOM List in force at the end of the reference period.
•Variable 251002 (Production under sub-contracted operations): (except for industrial services) national currency and quantity as defined in the PRODCOM List in force at the end of the reference period.
• Variable 251003 (Actual production): quantity as defined in the PRODCOM List in force at the end of the reference period. The survey sample for reference year 2023 is little less than 42,000 enterprises with approximately 65,000 local units, covering about 230 industries, derived from sub-sectors of the total manufacturing sector, and about 4,000 products (that include also industrial services). With the entry into force of the EBS (EU) regulation on European business statistics, the inclusion of the KAU (Kind of activity unit) statistical unit in the Registers managed by the National Statistical Institutes becomes mandatory. Consequently, this unit of analysis is also adopted in the PRODCOM survey.
30 August 2024
From 2021 onwards, the following information is requested for each variable:
The sold production (including industrial services) defined as sold (invoiced) production, which may be carried out under the primary or secondary activities of the enterprise. It includes production sold (invoiced) between different kind-of-activity units belonging to the same enterprise. The production under sub-contracted operations, which has been sold (invoiced) to the principal. The production may be carried out under the primary or secondary activities of the enterprise. The actual production includes any production carried out during the reference period and within the economic territory. With regard to subcontracting relationships we do not have the information of who is the owner of the goods.
Generally, the reporting unit is the Legal Unit as a proxy for the enterprise; the observation unit is Local Unit(s). From 2021 onwards the reporting unit is the KAU. However, we are using profiling in order to follow some dominant enterprises.
The survey population is made up of all industrial enterprises in NACE 07-33 (excepting 09 and 19) and 3832, with more than 20 employees and a sample of small industrial enterprises (from 3 to 19 employees).The population is selected every year from Statistical Business Register.
Italy - whole national territory.
Year 2023
For reference year 2023, the response rate is 56,9% of all enterprises surveyed.
The main causes of non-response are: misclassification of economic activity (activities outside of sections B and C, NACE Rev.2), inactive, merger, split off, unreachable unit.
After checking if the company has stopped or suspended its activity during the reference year, the non-responses were estimated using the previous year data and data from other sources. In case of partial responses, that concerns basically missing volume data, estimates for missing volumes are done by using the median unit price from the other respondent companies.
For Volume: Physical measurement unit required by Prodcom list, specific for each product.
For Value: Thousand Euro (excluding VAT and excise duties).
Steps of data compilation in PRODCOM statistics of Italy:
1. Error detection based on consistency edits
2. Manual error correction forcing wrong values on the base of deterministic rules as reported in a manual or document
3. Manual error correction forcing wrong values on the base of previous correct data of the same survey
4. Manual error correction based on reporting unit follow-up
5. Manual coding assisted by paper lists (manuals, listings) or computer screens
6. Data entry by Istat section in charge of the survey
7. Splitting questionnaire according to various record types
8. Deterministic error and outlier detection and imputation based on deterministic rules (IF-THEN)
9. Error and outlier detection and imputation by clerical reviewers based on experience
10. Error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
11. Error and outlier detection and imputation by clerical reviewers based on questionnaire review
12. Error and outlier detection and imputation by clerical reviewers based on respondent follow-up
13. Deterministic error and outlier detection and imputation by clerical reviewers based on experience
14. Deterministic error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
15. Deterministic error and outlier detection and imputation by clerical reviewers based on data from other surveys or sources
16. Deterministic error and outlier detection and imputation by clerical reviewers based on questionnaire review
17. Deterministic error and outlier detection and imputation by clerical reviewers based on respondent follow-up
18. Deterministic error and outlier detection and imputation by clerical reviewers based on the available auxiliary variables
19. Estimate calculation
The survey sample is selected from Statistical Business Register 2021. PRODCOM estimates cover Italy’s registered businesses for 90% (we made census for enterprises with 20 or more employees, which totals over 30.000 units).
For the enterprises with less than 20 employees we made a sample of over 10.000 units. VAT and turnover is classified to NACE divisions 07-33 and 38. Businesses are selected by employment size, industry classification and geographical breakdowns, using a stratified random sample. Additional constraints are built into the sample design so to decrease the form filling response burden on smaller businesses. The number to be selected has been calculated to minimize the burden of sampling, without compromising the calculation of final estimates.
The sampling method is based on an equiprobabilistic selection in each stratum obtained by the combination of two three variables: class of employees, class of economic activity and geographical repartitions. The variables used to evaluate the sample size in each stratum are: turnover and number of employees. The determination of sample size and its allocation between the stratum occurred requiring that the expected error in the estimated total (error expressed in terms of coefficient of variation) of the variables considered were less than 5 percent in each domain of study represented by the class of economic activity. The allocation of the sample in the layers was achieved by adopting a multivariate allocation algorithm proposed by Bethel, which is a generalization of the method of Neymann for only one variable, and that allows to minimize the sample size so as to satisfy the constraints precision of the estimates for each of the variables considered of interest, relative to the domain estimation.
Data are disseminated at national and at European level on yearly basis.
Data are disseminated at national level Yearly.
Data are provided to Eurostat Yearly within t+6 months.
Data are comparable with those of other European Union (EU) countries.
All EU countries produce PRODCOM statistics, which is the system for the collection and dissemination of statistics of industrial goods based on the PRODCOM List (8-digit codes).
In Italy, the List of Products used is based on the PRODCOM List with Italian codes added according to the interest of users.
With implementation of the EBS Regulation, a break of series due to the fact that the production subject to subcontracted operations was collected separately from own-account production from the year 2021.
From a NACE classification perspective, the introduction of NACE Rev. 2 in 2008 resulted in a break in series.
