Statistics on the production of manufactured goods (prom)

National Reference Metadata in Euro SDMX Metadata Structure (ESMS)

Compiling agency: Statistical Office of the Republic of Slovenia


Eurostat metadata
Reference metadata
1. Contact
2. Metadata update
3. Statistical presentation
4. Unit of measure
5. Reference Period
6. Institutional Mandate
7. Confidentiality
8. Release policy
9. Frequency of dissemination
10. Accessibility and clarity
11. Quality management
12. Relevance
13. Accuracy
14. Timeliness and punctuality
15. Coherence and comparability
16. Cost and Burden
17. Data revision
18. Statistical processing
19. Comment
Related Metadata
Annexes (including footnotes)
 



For any question on data and metadata, please contact: Eurostat user support

Download


1. Contact Top
1.1. Contact organisation

Statistical Office of the Republic of Slovenia

1.2. Contact organisation unit

Business statistics, Short-Term Business Statistics

1.5. Contact mail address

Litostrojska 54, SI-1000 Ljubljana


2. Metadata update Top
2.1. Metadata last certified 29/08/2023
2.2. Metadata last posted 29/08/2023
2.3. Metadata last update 29/08/2023


3. Statistical presentation Top
3.1. Data description

The statistics on the production of manufactured goods and industrial services carried out by enterprises on the national territory of Slovenia.

The survey is conducted by the Statistical Office of the Republic of Slovenia (hereinafter SURS). 

Data available includes:

  • actual production (in volume),
  • sold production in volume for goods and services produced on own account,
  • sold production in value for goods and services produced on own account,
  • the volume of the production under sub-contracted operations,
  • the value of the production of subcontractors.
3.2. Classification system

Data are collected and published based on the Nomenclature of Industrial Products (NIP), which is the Slovene version of the European Prodcom List. NIP is published on our website: Classifications and Code Lists: Economic Classifications (in Slovene only). NIP is based on the entire list of headings of the Prodcom List for the relevant year. The 8-digit headings of Prodcom List are for national purposes extended with one more digit, the NIP therefore consists of 9 digit. The 9th digit enables more detailed national division, if necessary because of specialties of Slovenian industry.

Data on sold production in value are published also according to the Classification of Products by Activity (CPA).

3.3. Coverage - sector

The survey covers enterprises and establishments that, according to the Standard Classification of Activities (NACE Rev.2), perform one or more activities from sections "Mining and quarrying" (B), "Manufacturing" (C) and class 38.32 Recovery of sorted materials. From section Mining and quarrying (B) divisions 05 Mining of coal and lignite, 06 Extraction of crude petroleum and 09 Mining support service activities are excluded.

3.4. Statistical concepts and definitions

Actual production of industrial products means the quantity of products produced in the territory of Slovenia by the enterprises settled in the territory of Slovenia, in the observed year, irrespective of whether the products are intended for sale, further production or stocks. Data on production include capitalized production and products given to the employees in kind. Products are made either from the producer's own material (production on own account) or from the principal's material, when the production is carried out by a sub-contractor under sub-contracted operations. The measurement unit, such as kilograms, square metres etc. used to indicate the volume of goods produced.

Production under sub-contracted operations is the quantity of products produced in the territory of Slovenia by an enterprise established in the territory of Slovenia in the reference year under a subcontracting relationship. A sub-contractor carries out the production where the raw materials and products are owned by the principal. 

Sold production in value of industrial products and services is the invoiced value of the ex-works selling price which includes also packaging costs even if they are charged separately, compensations and subsidies which refer to produced and sold quantities or values of products and services. Data on sales value include capitalized production and the value of products given to the employees in kind. However, the sales value does not include the value added tax, similar deductible taxes and duties, separately charged transport costs or discounts granted to customers. The data on sold production in value include values of the sold production produced on own account (including industrial services). The value is expressed in national currency (EUR). Until 2021, sold production in value include values of the sold products produced on own account (including industrial services) and the value of the products produced from the principal's materials under the subcontracting relationship. The value of the products produced from the principal's materials under the subcontracting relationship was equal to the market value of the products.

Sold production in volume of products is the quantity of individual products sold in the observed year, irrespective of whether products were produced in the observed year or before (sale of stocks). Data on sales volume include capitalized production and the volume of products given to the employees in kind. The data on sold production in quantity include quantities of the sold production produced on own account and is expressed in the unit of measurement prescribed in the NIP classification.The volume is expressed in a measurement unit specified for each product in NIP. Until 2021, sold production in volume include quantities of the sold products produced on own account (including industrial services) and the quantities of the products produced from the principal's materials under the subcontracting relationship. 

The value of the production of subcontractors contains values equal to the fee received by the enterprise as a subcontractor established in the territory of Slovenia for the production of products made from the principal's raw materials and sold to the principal under the subcontracting relationship.

3.5. Statistical unit

The observation unit is local kind of activity unit (LKAU). For the purpose of legislation the results are compiled at kind of activity unit (KAU) level.

3.6. Statistical population

The sample framework covers all legal forms of local kind of activity units (LKAUs) which, as their main activity, carry out one or more activities in section Mining and quarrying (B) and Manufacturing (C) and class 38.32 Recovery of sorted materials of the Standard Classification of Activities (NACE Rev.2) with 20 or more persons in paid employment and exceptionally some smaller enterprises. The sample frame consists of about 23,000 units, and about 3,000 units are selected to participate in the survey. The units were selected on the basis of threshold coverage.

3.7. Reference area

National territory of the Republic of Slovenia. No regional breakdown.

3.8. Coverage - Time

The observation period is a calendar year. 

A comparable time series, calculated according to the described methodology, is available from 1994 on.

Data starting from 2011 can be found in SiStat Database. Older data are not published.

3.9. Base period

Not applicable.


4. Unit of measure Top

For volumes unit of measure is specific to each product code and can be seen in the Nomenclature of industrial products (NIP), published in Slovene on SURS' website.

Values are in thousands EUR.


5. Reference Period Top

Year 2022.


6. Institutional Mandate Top
6.1. Institutional Mandate - legal acts and other agreements

This survey is carried out in compliance with following regulations:

  • at EU level: 
    • 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 (CELEX: 32019R2152)
    • 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 (CELEX: 32020R1197)
    • Commission Regulation for the PRODCOM List, valid for the reference year; for 2022 is valid Commission Regulation (EU) 2022/2552 of 12 December 2022 establishing the Prodcom list of industrial products provided for by Council Regulation (EEC) No 3924/91 (CELEX: 32019R1933)
  • at National level: National Statistics Act (OJ RS, No. 45/95 and 9/2001) and the Annual Program of Statistical Surveys (only in Slovene).
6.2. Institutional Mandate - data sharing

Not applicable.


7. Confidentiality Top
7.1. Confidentiality - policy

All data collected and published by SURS are governed by the National Statistics Act (OJ) RS No. 45/95 and (No. 9/01). The National Statistics Act states that statistics may be published in aggregate form only and are equally accessible to users. By way of exception, data may also be published individually:

  • upon written consent of the reporting unit as regards publication of the data in such a way;
  • if data have been collected from public (generally accessible) data collections (records, registers, databases, etc.);
  • if data are published in such a way that the reporting unit involved cannot be identified.
7.2. Confidentiality - data treatment

The p%-rule and the threshold rule for disclosure control on all the product groups are used. The algorithm used for secondary suppression is Modular from Tau-Argus. The results from the disclosure control on the product codes are marked up (flagged) in the data transmission to Eurostat.


8. Release policy Top
8.1. Release calendar

Provisional data on the reference year (t) are published in June (t + 6 months).

Final data on on the reference year (t) are published in September (t + 9 months).

Dates of publication of provisional and final data are announced in the release calendar. The release calendar is prepared in advance in the current year for the next calendar year.

8.2. Release calendar access

The calendar is placed on the official statistics website of SURS.

8.3. Release policy - user access

The most important information channel for data publication is the official statistics website of SURS. New data are realised according to announced dates in released calendar at 10.30 a.m. None of the users have a prior access to the data. The simultaneous issue of the First Release is ensured. Same release policy applied to national release is applied to transmissions to Eurostat.

All data are available free of charge, without licence limitations. Our copyright policy only requires that SURS is stated as the source of data. 
The subscribe to weekly notifications of upcoming releases is available. The same is true of a brief daily notification about newly released data. 


9. Frequency of dissemination Top

Prodcom data are disseminated annualy; 6 months after the end of the reference year as a provisional data and 9 months after the end of the reference year as a final data.


10. Accessibility and clarity Top
10.1. Dissemination format - News release

Provisional data are issued with a notice called First Release (Production and sold production of industrial products and services)

Detailed provisional data are published in the SiStat Database and transsmited to Eurostat at the same time.

Final data are published in the SiStat database and transsmited to Eurostat at the same time.

10.2. Dissemination format - Publications

Not available.

10.3. Dissemination format - online database

 SiStat Database

10.4. Dissemination format - microdata access

The entities entitled to obtain the statistically protected microdata include the registered research institutions, registered researchers, and the researchers of government offices. Basic instructions concerning the access and the use of statistically protected microdata are availabe on the SURS website.

10.5. Dissemination format - other

The Prodcom data are transmitted to Eurostat according to EU legislation in order to compile EU aggregates and for the dissemination of national data in Eurostat’s dissemination database.

Annual submission of data to national institutions (i.e. Chamber of Commerce and Industry, Forestry Institute,...) and to international institutions (i.e. FAO, UN,...)

Internal submission of data to national accounts, price statistics and environmental statistics.

Other special required data in accordance to the special needs of the applicants.

Information Centre of SURS.

10.6. Documentation on methodology

Methodological explanations for Prodcom data are availabe on SURS' website: Methodological explanations, Theme: Industry, Sub-theme: Production and Turnover in Industry. There is an automatic link to the methodological explanations in the electronic version of the First Relase and at the SiStat Database for the data set on Annual production for industrial products and services.

Also questionnaire is available on-line (only in Slovene): Questionnaires for statistical surveys, Theme: Industry, Sub-theme: Production and Turnover in Industry. 

A number of internal papers and notes exists on methodological aspects such as data validation/editing process, aggregation, coverage and sampling etc.

10.7. Quality management - documentation

Quality reports are available on-line: The quality reports for statistical surveys, Theme: Industry, Sub-theme: Production and Turnover in Industry.


11. Quality management Top
11.1. Quality assurance

SURS in performing its tasks follows the general principles of quality management, the European Statistics Code of Practice and the Fundamental Principles of Official Statistics. The principles are more in detail presented in the Quality Statement of the Statistical Office of the Republic of Slovenia.

The quality of statistical data used to be dealt with mostly in connection with data accuracy in the narrow sense (as coherence between statistical data and exact values). In the last decade the statistical profession has made great progress towards broader understanding of the quality of statistical data. Quality is now dealt with in terms of different quality dimensions: relevance, accuracy of estimates, timeliness and punctuality of publication, accessibility and clarity of information, comparability of statistics and coherence of results. SURS regularly publishes reports on the quality of statistical surveys, which contain detailed descriptions of individual statistical surveys regarding all quality dimensions. Quality reports also contain the values of quality indicators, i.e. numerical values of achieved quality levels for individual quality components.
 
Before sending data to Eurostat, SURS apply a number of micro and macro plausibility checks at different aggregation levels, between historic and current data and with other sources.
11.2. Quality management - assessment

Data on Prodcom statistics are compiled in compliance with methodological requirements and standards.


12. Relevance Top
12.1. Relevance - User Needs

The main users of Prodcom data are:

  • institutions at European level (DGs, Eurostat),
  • multi-national organizations (UN, FAO, International road federation, US Aluminium Association),
  • the national goverment and their institutions (Slovenian environment agency, Ministry for agriculture, forestry and food, The National Institute of Public Health),
  • non-governamental institutions at national level (Slovenian forestry institute, Chamber of Commerce and Industry of Slovenia, Agricultural Institute of Slovenia, Institut Jožef Stefan),
  • the media,
  • researchers and students,
  • the enterprises,
  • internal users, like national accounts, price statistics, agricultural statistics.
12.2. Relevance - User Satisfaction

Communication with users takes place at meetings of the Business Statistics Advisory Commitee, chaired by a representative of the Faculty of Economics, University of Ljubljana. The representatives of Slovenian Chamber of Commerce, Chamber of Commerce and Industry of Slovenia, School of Economics and Business, University of Ljubljana, Ministry for Economic Development and Technology, Institute of Macroeconomic Analysis and Development, and National accounts section are members of the commitee. The meetings take place every 18 months, the last meeting was on 1 March 2023. More about the work of the commitee is available at (in Slovene only):

https://www.stat.si/statweb/NationalStatistics/AdvCommitteesDescription/97.

Additionally, communication with users is managed through the Information centre, which helps them search for adequate statistical data. User iniciatives are managed by the responsible methodologist, who also manages direct communication with the key users.

There is a need on the part of key users for data that are already being collected through Prodcom survey, but cannot be published due to confidentiality. Data confidentiality is a problem for data users, as approximately 70% of the data at the level of NIP code are protected due to confidentiality.

SURS measured general user satisfaction for the last time in 2022. Respondents assessed general satisfaction with the SURS products and services with the average score of 8.3 (on a scale from 1 – disagree completely to 10 – agree completely).

12.3. Completeness

All statistics required by the current EU legislation are provided.


13. Accuracy Top
13.1. Accuracy - overall

The survey is based on census with cut-off treshold, all companies in the relevant NACE Rev.2 groups that have at least 20 employees (in some cases less) are surveyed. The basis for sample frame is Statistical Business Register. The sample frame consists of about 23,000 units, and about 3,000 units are selected to participate in the survey.

13.2. Sampling error

The survey is based on census with cut-off threshold, so there is no sampling error.

13.3. Non-sampling error

Unit non-response rate

Unit non-response rate shows the percentage of units that did not provide us with any information. The indicator is determined by the ratio between the number of units for which we have received no information and the number of eligible units. At the time of the first release unit non-response rate was of 17.4% (size-weighted of 8.2% - a number of employees is used as a weight). The questionnaire keep being opened two months after the first release of proivisional data. This way new questionnaires are recorded after the first publication of the results and the non-response rate decreases slightly. During the whole data collection process the response rate is checked and attention is paid to get the data of the key respondent units.

In order to ensure the highest possible level of response, we offer methodological and technical assistance in completing the questionnaire on the toll-free telephone number. We send a »do not forget« reminder before the reporting deadline and a reminder after the due date, where we inform them about the legal obligation to participate in the SURS's survey. For the units that still do not respond, telephone follow-up is carried out. If the unit still does not respond, the non-response is replaced by statistical estimates of imputation.

Item non-response rate

The survey does not include item non-response, because the WEB questionnaire was upgraded in 2022 in a way that units that completed the questionnaire have to answer all key questions.

Imputation rate

Table 13.1: Imputation rate (%) by key variables

 

2022

Actual production

11.8

Sold production in volume

11.8

Sold production in value

11.8

 

Imputation rate for new varaibles; production of subcontractors and the value of the subcontracted production are not calculated because the data for new variables are derived from the data on total production/sold production and type of production in the dissemination phase and are not part of the database as such. Thefore the imputation rate for new variables are part of the imputation rate for actual production and sold production in value.

Coverage bias

In order to assess the impact of exclusion of part of the population from the survey, we calculate coverage bias. Coverage bias measures the error in a statistical estimate caused by the fact that a part of the population was left out of the observation on purpose. Bias due to other factors, such as non-response, measurement errors, processing errors, etc., is not estimated. We estimated the value of sales for the whole population and then for the part of the population that was included in the survey. In both cases the value of sales was calculated on the basis of the annual reports of the enterprises.

Table 13.2: Coverage bias, 2022

 

Domain name

Domain value

Statistics

Bias

Total

 

values

0.93

NACE Rev.2

08

values

0.89

NACE Rev.2

10

values

0.92

NACE Rev.2

11

values

0.93

NACE Rev.2

13

values

0.92

NACE Rev.2

14

values

0.87

NACE Rev.2

15

values

0.90

NACE Rev.2

16

values

0.87

NACE Rev.2

17

values

0.94

NACE Rev.2

18

values

0.81

NACE Rev.2

20

values

0.93

NACE Rev.2

21

values

1,00

NACE Rev.2

22

values

0.91

NACE Rev.2

23

values

0.92

NACE Rev.2

24

values

0.99

NACE Rev.2

25

values

0.90

NACE Rev.2

26

values

0.92

NACE Rev.2

27

values

0.97

NACE Rev.2

28

values

0.90

NACE Rev.2

29

values

0.96

NACE Rev.2

30

values

0.83

NACE Rev.2

31

values

0.81

NACE Rev.2

32

values

0.90

NACE Rev.2

33

values

0.75


The control system for detecting measurement errors in data are based on computer-aided automatic procedures. The so-called logical controls represent a set of controls that relate to the adequacy of the data of an individual unit for the variables collected by the questionnaire. Adequacy is checked in various ways, e.g. taking into account the size of previously transmitted data (e.g. the production of a certain product varies by 30% from the production of the same product in the previous year), the relation between monitored variables (e.g. the value of sales in foreign markets is greater than the total sales), etc. For each variable of the questionnaire, a list of conditions is prepared that must be met in order to be able to conclude that the value of the variable is within logical, acceptable limits.Measurement error

Controls are marked as an error or as a warning. Checks labelled as errors represent an almost certain reporting error, and controls labeled as a warning point out that the reported data deviate but their values are not impossible. Errors should, as a rule, be rectified before handing over the material to the further stage of statistical processes, while warnings may remain. Logical controls are set and are the same for all units, only a few value data controls set the limits for large units more narrowly than for small ones. If we suspect an error, we contact the reporting unit and ask for relevant, corrected data, or further explanation as to why the variable exceeded the predicted limits. In the event of an error, the data are corrected manually.

Each clarification provided by the reporting units is systematically recorded, so that the units do not have to be contacted repeatedly and that the potential outliers in the published results can be explained.

The main reasons for measurement errors are:

  • the person completing the questionnaire misread or skipped reading the instructions. The variable definition was therefore not taken into account. A different person completing the questionnaire for the same reporting unit could interpret the questionnaire or the data for the unit differently;
  • the person completing the questionnaire choose wrong product codes or do not report complete set of products. Prodcom codes reported last year are pre-printed on the questionnaire. There is a weakness the enterprises only use the pre-printed codes, even though the new products were produced.
  • reporting in wrong measurement unit or difficulties in conversion from one measurement unit to another;
  • a change in business demography (i.e. merger, division) and the reporting unit hasn't updated its records yet.


14. Timeliness and punctuality Top
14.1. Timeliness

Preliminary data are available 6 months after the end of reference year and final data are available 9 months after the end of reference year.

Data are published on national level and submitted to Eurostat for publication at the same time.

Table 14.1: Timeliness of the first results

Reference period

2022

Date of publishing

21 June 2023

Time lag 

t+6

14.2. Punctuality

Table 14.2:Punctuality of the first results

Reference period

2022

Announced date

21 June 2023

Publishing date

21 June 2023

Difference

0

Punctuality of the first release is calculated as the difference between the date of announcement and the date of publication. 


15. Coherence and comparability Top
15.1. Comparability - geographical

The data on Prodcom statistics are available for a national territory of Slovenia as a single geografic area. 

The data submitted to Eurostat fully compliant with EU requirements.

15.2. Comparability - over time

Until 1994, the basis for data on industrial production and sales was the National Classification of Activities (EKD). Since 1994 the data have been collected according to the Nomenclature of Industrial Products (NIP), which changes over the years. There is no correspondence table between EKD and NIP. There is only a correspondence between the 6-digit EKD code and the 2-digit SKD code. There is also no correspondence between NIP 95 and NIP 97.

A comparable time series, calculated according to the described methodology, is available from 1995 on. In 2005, the Prodcom methodology was changed, so enterprises registered according to the 2002 Standard Classification of Activities in section E - Electricity, gas and water supply and subsections CA - Mining and quarrying of energy producing materials and DF - Manufacture of coke, refined petroleum products and nuclear fuel are no longer covered. According to the 2008 Standard Classification of Activities these enterprises are registered in section D – Electricity, gas, steam and air conditioning supply and divisions 05 - Mining of coal and lignite, 06 - Extraction of crude petroleum and natural gas and 09 - Mining support service activities of section B - Mining and quarrying. In 2021, for the first time, the survey covered enterprises classified under class 38.32 Recovery of sorted materials.

The method of monitoring the value of sold production in the case of production under sub-contracted operations was also changed. Before 2005 the value of sold production under sub-contracted operations was evaluated as the value paid by the principal for the service performed (a fee), and from 2005 on as the market value of the product, i.e. the value that the product would achieve on the market.

In 2021, the method of monitoring the value of sold production in the case of production under sub-contracted operations was changed; the value of sold production under sub-contracted operations is again evaluated as the value paid by the principal for the service performed (a fee). In addition, according to the new methodology, the value of payments received for subcontracted production is shown separately from the value of sales of products produced from own materials and the value of industrial services rendered. And only sales of manufactured goods and industrial services is included in the data on the quantity of goods sold and services provided. Changes in methodology have led to a 4% increase in the value of sold production under sub-contracted operations at the level of the whole industry in 2005 and a decrease of around the same amount in 2021, while at lower levels the share fluctuates depending on the share of subcontracted production in each industry.

15.3. Coherence - cross domain

We have compared data on total sold production in value from the Prodcom annual survey and nominal turnover indices from the STS monthly survey. The monthly STS survey monitors the turnover and new orders' trends and value of stocks in manufacture sector.

Table 15.1: Coherence with the reference survey

c - confidential

The differences between the Prodcom survey indices and the STS survey indices are mainly the results of different methodologies. In contrast to the STS monthly survey, in the Prodcom annual survey more reporting units are included and data are collected for the previous year, while in the monthly survey data are collected on the monthly basis for the current year. In addition, the differences occur due to the different methodologies of reporting the value of the subcontractor operations. In the STS survey, as in the Prodcom survey from 2021 on, subcontractors report a fee received for a service, but in STS survey the principal report the turnover from the products sold, while in the Prodcom survey the principal do not report anything.

15.4. Coherence - internal

Data are internally coherent. Each level of aggregated data is compiled directly from micro data.


16. Cost and Burden Top

Table 16.1: Survey costs at the statistical office

Reference period

2022

Number of working hours spent

5,186

Number of reporting units that had to fill in questionnaires

2,988

Survey period

annual

Number of questionnaires per year (total)

2,988


Table 16.2: Burden of the reporting units

Reference period

2022

Number of reporting units that submitted the data

2,426

Annual number of questionnaires per unit

1

Time spent to fill in a questionnaire (hours)

2.0

Total time spent (hours)

4,852


17. Data revision Top
17.1. Data revision - policy

 Methodological explanations about data revision at SURS.

 The same revision policy is applied nationally and in transmission to Eurostat.

17.2. Data revision - practice

Provisional data are published no later than the end of June for the previous year. Final data are based on more complete answers about the phenomenon and analyses and are published at the end of September for the previous year at the latest. Only the latest data (for the previous year) are being revised, older data that were published a year before or earlier are final and are not revised anymore. Data revision is planned.

The differences between provisional and final data are relatively small or almost zero. The main purpose of the annual industry is data on production and sales at the level of 9-digit NIP codes. Users are mainly interested in data on the production of an individual product at the NIP code level or smaller aggregates for two or three NIP codes together and not aggregated data at higher levels. Individual errors can have a major impact on data quality at such a detailed level. Provisional data are available at the end of June and final data are available at the end of September. Where accuracy of data needs to be verified with the reporting unit, the detailed control takes place in the time between publication of provisional and final data. The reason for the difference between provisional and final data is often either incorrectly classified products or reporting in the wrong unit of measure.


18. Statistical processing Top
18.1. Source data

Prodcom data are collected by the SURS. The data are collected through statistical survey. 

The survey is conducted by questionnaire, targeted all enterprises and establishments with 20 or more persons in paid employment and exceptionally some smaller enterprises. In some cases we have to include also smaller enterprises. Although we go below the 20 employees line, we still have problems reaching 90% of production even at the 2-digit (division) level of NACE Rev.2. 

The source for reported data is the accounting documentation of enterprises and only exceptionally their estimates.

Data for the survey are not obtained from administrative sources.

 

18.2. Frequency of data collection

Data are collected annualy.

18.3. Data collection

The data are collected via web application for electronic reporting of data and via post for paper version of the questionnaire. Web questionnaire opens in February. Almost 99% of all units that respond to the questionnaire provide us with data via E-questionnaire. As we want to switch units to electronic reporting, a paper questionnaire is sent to reporting units only on their request. The enterprises have to provide filled in questionnaires by the end of February. Data from questionnaires are then being transferred to input database. Through March, April and May the data are being checked in logic control. Through all this time claiming takes place and additional questionnaires come in. We also contact reporting units for additional explanation on their data. 

In data collection phase SURS uses the system of key respondents. Key respondents are the most influential units that are pre-selected to be treated differently in the data collection and data editing phases. During the data collection phase we make a lot of effort to obtain complete and accurate response from them. In Prodcom survey the “do not forget” reminder is sent out five days before the due date, the first reminder is sent 7 days after the due date and the second reminder is sent 14 days after the due date. Shortly after the second reminder it is checked if the key units have answered the questionnaire. If not, telephone follow-up of the key units is carried out by the Call Centre analysts. Key units are reminded to answer the questionnaire and send it to SURS. With the first reminder the units are kindly asked to participate in the survey. In the second reminder the units are informed about their legal obligation to participate in SURS’s surveys. With every reminder also a link to the online questionnaire and a phone number are attached, where units can participate in the survey. The units not responding are not fined.

18.4. Data validation

In data validation is implemented selective editing approach. Implementation of selective editing caused a significant change in the whole statistical process; a new application for manual data editing (contrary to previously used Blaise application) operates directly on the input database. 

After the initial data validation (duplicates, missing identifiers, etc.), the variables are divided into two (disjoint) sets. One set of variables is edited automatically and the other is edited manually. Here manually means that we re-contact the units and interview them again to clear out the errors that occur in the data. If data are very different from those reported by the same company for the last period, the respondents will be prompted to check data again and to provide an explanation for the change. The micro data are validated through a number of controls that check against last period as well as against other companies reporting under the same codes (unit price control). Since a part of the variables is edited automatically, the number of units that have to be re-contacted is essentially reduced.

The aggregated statistics after compilation are compared over the previous year data and also checked against other sources, especially the STS  and SBS data (a turnover index). 

18.5. Data compilation

After data validation phase imputation procedures are run. The term data imputation refers to all procedures in which missing or inaccurate values detected in the data editing process are replaced by statistical estimates. The data imputation process for estimation of missing values uses deterministic methods (estimated value is calculated by an analytical procedure using an appropriate deterministic function) and stochastic methods (estimated value is calculated by a procedure using a probability mechanism).

The following methods were used:

  • Logical imputations; imputed value is calculated from the values of other variables.
  • Historical data imputations; imputed value is calculated as the average trend of the auxiliary varables, multiplied by the historical value of the variable.
  • Donor imputations; imputed value is taken from another unit (donor). Donor is selected  in such way that the donor is similar to the recipient with respect to the (one or more) matching variable(s) (for example number of employees, NACE Rev.2).

The units are in the survey selected on the basis of threshold coverage; consequently grossing and weighting techniques are not used. All results are presented in the form of absolute figures. Data on actual production and sold production in volume are shown at the level of the product or service. The basis is the Nomenclature of Industrial Products (NIP). Due to the different units of measurement, quantitative data can not be shown at higher levels. Data on the value of sold production are shown at the level of the product (NIP) as well as at higher levels (section, division, group and class of CPA).



18.6. Adjustment

No other adjustment is needed.


19. Comment Top

No comment.


Related metadata Top


Annexes Top