# Documentation of statistics Statistics on service industry commodities

16.8.2022 valid documentation

## Basic data of the statistics

### Data description

Statistics on service industry commodities describe the structure of turnover and the service selection in industries producing services. The statistics are based on Regulation (EC) No 295/2008 of the European Parliament and of the Council concerning structural business statistics. It defines the branches of industry involved, the data content of the statistics and the service item classification used.

### Statistical population

According to the Regulation, the population of the statistics covers all enterprises employing at least 20 persons in the industries mentioned above. However, in many industries the lower limit of 20 persons results in the group of enterprises only covering a small share of the industry's turnover. In addition, small and large enterprises specialise in the production of certain services in some industries. For this reason, the lower limit of the population is five persons in certain industries. The objective in defining the industry-specific employee limit is that the frame covers 80 per cent of the industry's turnover and produces a representative picture of the structure of the industry.

The frame is stratified by industry and size category according to employees. Random sampling is utilised in those strata from which it is not necessary to select all enterprises into the inquiry group. All enterprises employing at least 100 persons are selected to the inquiry group.

The statistics do not describe the production of business services fully, because due to the above-mentioned employee limitations, the smallest enterprises in the industries are not included in the population of the statistics.

A significant factor affecting the reliability of the statistics is the ability of the respondents to break down turnover according to the CPA classification or export areas. Efforts have been made to manage this risk by specifying the classification and the descriptions of the categories.

### Statistical unit

Data from the statistics on service industry commodities are published on the basis of legal units. The data are sent to Eurostat based on enterprise units. According to Statistics Finland's definition, an enterprise is an independent unit in its decision-making that produces goods and/or services for sale on the market. An enterprise may be formed of one or several legal units. If an enterprise has been formed of several legal units, transactions between the units have been eliminated.

Legal units include:
•    persons carrying on a business, trade or profession in their own name or in the name of a registered firm,
•    legal persons (e.g. limited company, limited partnership, general partnership, cooperative, savings bank and economic association),
•    market-based enterprises.

The legal unit is an administrative reporting unit and a reporting unit for its own data collections.

### Unit of measure

The units of measure used in the database tables of the statistics are EUR 1,000 and per cent.

### Base period

The statistics on service industry commodities are not index-based statistics.

The data are specific for each year.

### Reference period

The reference period of the statistics on service industry commodities is the calendar year. The data relate to the accounting period that ended in the statistical reference year, so the accounting period does not always follow the calendar year.

Extended and shortened accounting periods are converted to correspond to 12 months.

### Reference area

The statistics on service industry commodities are published on the level of the whole country.

The data also cover activities that have been carried out outside the borders of Finland but that are included in the enterprise's financial statement made in Finland. Thus, the turnover of the statistics may also include such as goods sent abroad for processing and sales of goods and services from abroad to abroad.

### Sector coverage

Statistics on service industry commodities cover annually the following branches of industry:

IT services, advertising services, employment services, legal services, accounting, bookkeeping and auditing services, management consultancy services, architectural and engineering services, technical testing and analysis, and market research services.

According to the regulation, the population of the statistics covers all enterprises with at least 20 employees in the above-mentioned branches of industry. However, in some branches the lower limit of 20 employees means that the group of enterprises covers only a small part of the total turnover of the industry, thus not producing a representative view of the structure of the industry (larger enterprises focus on services different from smaller ones). For this reason the limit has been lowered in some industries to cover enterprises employing five persons.

### Time coverage

Annual data on the statistics on service industry commodities are available on the home page of the statistics and in Statistics Finland's StatFin database from 2009 to 2019.

The data are not are revised.

### Frequency of dissemination

Data on the statistics on service industry commodities are published once a year on Statistics Finland's website. The collected data relate to the accounting period that ended in the statistical reference year. The released data are final annual data and preliminary data are not published. In regular production the data collection takes place in the autumn following the statistical reference year and the time of publication is in December. Comparable data are available starting from the statistical reference year 2008.

## Accuracy, reliability and timeliness

### Overall accuracy

The main source of error is non-response.

The statistics on service industry commodities do not describe the production of services fully either, because due to the above-mentioned employee limits, the smallest enterprises in the industries are not included in the population of the statistics.

A significant factor affecting the reliability of the statistics is the ability of the respondents to break down turnover according to the CPA classification or export areas. Efforts have been made to manage this risk by specifying the classification and the descriptions of the categories.

### Timeliness

The statistics are published once a year. The collected data relate to the accounting period that ended in the statistical reference year. The released data are final annual data and preliminary data are not published. In regular production the data collection takes place in the spring following the statistical reference year and the time of publication is in the autumn.

### TP2

The final data of the statistics on service industry commodities are released annually within 12 months from the end of the statistical reference year. The release has taken place in December.

### Punctuality

There is no delay in publishing the data of the statistics on service industry commodities. The data have been published on the days indicated in the release calendar.

### Completeness

The statistics on service industry commodities cover annually the following branches of industry:
IT services, advertising services, employment services, legal services, accounting, bookkeeping and auditing services, management consultancy services, architectural and engineering services, technical testing and analysis, and market research services.

### Sampling error

Population

According to the definition of Regulation (EC) No 295/2008 of the European Parliament and of the Council concerning structural business statistics, the population of the business services statistics covers all enterprises employing at least 20 persons in certain industries providing business services. However, in many industries the lower limit of 20 persons would result in the population covering only a small share of the industry's turnover. In addition, small and large enterprises specialise in the production of certain services in some industries.
In the business services statistics the basis in defining the industry-specific employee limit is that the frame covers 80 per cent of the industry's turnover and produces a representative picture of the structure of the industry. The data are collected from enterprises employing either at least 5 or 20 persons depending on the industry. The turnover of enterprises employing under five persons is often so low that asking of data from these enterprises is not justified due to the response burden and the significance of the received data. The use of more than two lower limits of employees is neither justified for the clarity of the statistics.
The frame has been stratified to TOL 3, 4 or 5-digit level depending on the industry so that a representative group of enterprises can be obtained from all industries to the frame. The stratification is based on the fact that according to the Regulation on structural business statistics the business services statistics cover certain CPA categories. The CPA follows the TOL2008 classification, so when compiling the inquiry group it must be ensured that the respondent enterprises are obtained from all CPA categories according to the Regulation. The stratification has been made on a more detailed 5-digit level in cases where a service according to a certain CPA category is produced significantly in only one TOL 5-digit level industry.
In addition to the stratification by industry, the frame has been stratified by the number of personnel into four size categories: 5 to 19, 20 to 49, 50 to 99 and at least 100 persons. All strata based on the industry are divided into these size categories, which resulted in a total of 92 strata. The number of strata is affected by that not all size categories were represented in all strata by industry.

Enterprises employing at least 100 persons have been drawn into the inquiry group as certain. Due to the small size of the population, around 3,587 enterprises are drawn into the inquiry group from several industry strata. Random sampling is utilised in those strata for which it is not justified to select all enterprises into the inquiry group.

The enterprises are drawn by random sampling so that the sampling plan defines the number of enterprises to be drawn from the stratum. The probability proportional to size (PPS) method is used in the sampling. With this method the probability of selecting a unit grows as the unit's turnover grows. The aim is to get as large a share as possible of the turnover of the industry into the sample.

Calculation method

The effect of non-response is corrected by calculating the non-response correction coefficient for each stratum based on the numbers of the responding enterprises.

When the data are raised to the whole population, enterprises' turnover data are picked again from the same database of Statistics Finland that is used in the compilation of other business statistics. The enterprise-specific sample weight is multiplied by the stratum-specific non-response correction weight calculated in the first stage, which produces a preliminary weighting coefficient for the enterprise. The turnover data raised with this weighting coefficient are summed by stratum and the sums are compared with the stratum sums selected from the database. Based on these differences, the weighting coefficients are still revised so that the sums match.

### Coverage error

The data of the respondents are raised by stratum to correspond to the data of the population. Non-response and over-coverage and under-coverage are taken into account when forming estimates. Over-coverage includes enterprises classified in the wrong industry in the Business Register and enterprises whose closure has not been updated in the Business Register by the time the sample was drawn. In some individual cases the data of some enterprise not having responded may also be imputed.

## Comparability

### Comparability - geographical

The data cover the whole of Finland.

The data of the statistics on service industry commodities are based on the international industrial classification and the CPA2015 product classification.

The data sent to Eurostat are internationally comparable.

### Comparability - over time

A comparable time series on the data of the statistics on service industry commodities starts from 2008. However, changes in the coverage of strata must be taken into account in some industries.

Previously, the data on legal services, accounting, bookkeeping and auditing services, management consulting services, architectural and engineering services, technical testing and analysis services, and market research and public opinion polling services were collected every other year. From 2012 onwards, the data for these industries are also collected annually.

### Coherence - cross domain

Turnover data are not collected with the CPA classification in other sources. The frame for the data collection was drawn from Statistics Finland's Register of Enterprises and Establishments.

### Coherence - sub-annual and annual statistics

The statistics on service industry commodities are released once a year. The released data are final annual data and preliminary data are not published.

### Coherence -national accounts

The data of the statistics on service industry commodities are used as source data for the national accounts and the producer price index for services.

### Coherence - internal

The data of the statistics on service industry commodities are internally uniform.

## Source data and data collections

### Source data

Data by service commodity are not available from other sources, so the data must be asked directly from enterprises. The frame for the data collection was drawn from Statistics Finland's Register of Enterprises and Establishments. The data are based on data on legal units.

The business services inquiry collects information on the production of business services. Enterprises are asked about their turnover by service product category according to the CPA product classification, the breakdown of turnover between private and public sector customers, and the breakdown of turnover by customers’ country of location, divided into domestic, foreign within the EU area, and foreign outside the EU area.

According to the Regulation, the population of the statistics covers all enterprises employing at least 20 persons in the industries mentioned above. However, in many industries the lower limit of 20 persons results in the group of enterprises only covering a small share of the industry's turnover. In addition, small and large enterprises specialise in the production of certain services in some industries. For this reason, the lower limit of the population is five persons in certain industries. The objective in defining the industry-specific employee limit is that the frame covers 80 per cent of the industry's turnover and produces a representative picture of the structure of the industry.

The frame is stratified by industry and size category according to employees. Random sampling is utilised in those strata from which it is not necessary to select all enterprises into the inquiry group. All enterprises employing at least 100 persons are selected to the inquiry group. The data collection annually targets about 1,400 enterprises.

### Data collection

Data for Statistics Finland's own data collections are gathered with a web questionnaire. In addition, Statistics Finland has an agreement that the Tax Administration delivers the data of the business taxation data to Statistics Finland on the agreed dates.

The web questionnaire of the inquiry for statistics on service industry commodities has been tested in connection with its introduction and when changes have been made to it. Based on customer feedback, the instructions for the questionnaire have been specified.

The non-response of the inquiry is monitored yearly.

### Frequency of data collection

The data collection for the statistics on service industry commodities is carried out once a year.

### Cost and burden

The response burden was measured with an inquiry in 2021. Efforts have been made to minimise the response burden.

## Methods

### Data compilation

The content checks are used to examine possible response errors that cannot be detected with the data collection application checks. The data are raised to represent the population.

### Data validation

Content checks made to the data:

It is checked that the enterprise belongs to the correct industry by comparing the CPA category with the biggest turnover share to the enterprise's industry data on the TOL and CPA 3-digit and 4-digit levels. If the biggest CPA category answered by the enterprise does not correspond to the industry, these enterprises are revised in cooperation with the Register of Enterprises and Establishments.

Large enterprises whose turnover is recorded in full in one CPA category are examined. For these enterprises, it should be examined by means of the enterprise's web pages and direct contacts whether the enterprise actually concentrates on producing only one type of service or whether the turnover could be specified in more detail. In recent years, the focus has been on companies with a turnover of over EUR 10 million.

The relative distribution of CPA categories by industry is compared with the distributions of previous years. Possible significant deviations in relative shares are examined in more detail.

### A7

Calculation method

The effect of non-response is corrected by calculating the non-response correction coefficient for each stratum based on the numbers of the responding enterprises.

When the data are raised to the whole population, enterprises' turnover data are picked again from the same database of Statistics Finland that is used in the compilation of other business statistics. The enterprise-specific sample weight is multiplied by the stratum-specific non-response correction weight calculated in the first stage, which produces a preliminary weighting coefficient for the enterprise. The turnover data raised with this weighting coefficient are summed by stratum and the sums are compared with the stratum sums selected from the database. Based on these differences, the weighting coefficients are still revised so that the sums match.

<font color="#000000"><font face="Times New Roman"> </font></font>

### Documentation on methodology

A methodological description of the statistics on service industry commodities can be found on the home page of the statistics (in Finnish only).

## Principles and outlines

### Contact organisation

Statistics Finland

### Contact organisation unit

Economic Statistics

### Legal acts and other agreements

The compilation of statistics is guided by the general act of the national statistical service, the Statistics Act (280/2004, amend. 361/2013). Only the necessary data that are not available from administrative data sources are collected from data suppliers. The data are published so that no individual enterprise's data or development can be deduced from them.

The statistics on service industry commodities are based on Regulation (EC) No 295/2008 of the European Parliament and of the Council concerning structural business statistics, which defines the industries and data content covered by the statistics. The first reference year of the statistics is 2008.

The Regulation on statistical units, Regulation No 696/93 of the Council on statistical units, is applied to the statistics on business activities.

The Standard Industrial Classification 2008 is based on the European Union's common industrial classification, NACE Rev. 2, supplemented for national needs on the fifth level.

### Confidentiality - policy

The data protection of data collected for statistical purposes is absolutely guaranteed in accordance with the Statistics Act (280/2004), the Personal Data Act (532/1999) and the Act on the Openness of Government activities (621/1999), as well as the requirements of the EU's General Data Protection Regulation (2016/679). The data materials are protected at all stages of processing with the necessary physical and technical solutions. Statistics Finland has compiled detailed directions and instructions for confidential processing of the data. Employees have access only to the data essential for their duties. The premises where unit-level data are processed are not accessible to outsiders. Members of the personnel have signed a pledge of secrecy upon entering the service. Wilful breaching of data protection is punishable.

### Confidentiality - data treatment

Data on the statistics on service industry commodities are published on aggregate levels. The data of the statistics are published on Statistics Finland's website so that data on individual enterprises cannot be identified. In the statistics, the threshold value rule is used when assessing the disclosure risk of data for all industries/CPA categories. The protected data are suppressed. In primary suppression, data on an individual industry group are protected by suppression if the group has data on one or two enterprises. In secondary encryption, the data are suppressed so that the primarily protected data cannot be derived.

Enterprise-level data can be released encrypted for researchers' use in Statistics Finland's research services. Contact information for research services: http://www.stat.fi/tup/mikroaineistot/index_en.html

The information service follows the same protection rules.

The data sent to Eurostat that are subject to both primary and secondary encryption are indicated with codes.

### Release policy

Statistics Finland's release calendar lists in advance all the statistical data and publications to be released over the year. Statistical releases can be found under statistics-specific releases. Statistical data are released on the Internet at 8 am, unless otherwise indicated. The calendar is updated on weekdays. Statistics Finland's release calendar for the coming year is published every year in December. The dates of future releases can also be seen on the home page of the statistics.

### Data sharing

Data on the statistics on service industry commodities are published annually on the statistics' website and in the StatFin database.

The data are also submitted to the national accounts, the producer price index for services, the information service subject to charge and the research services.

In addition, industry-specific data on service items are supplied to Eurostat, which publishes the data on its web pages.

### Other

The industry-specific data of the statistics are delivered to Eurostat, which publishes data by industry on its own pages.

Industry-specific data are also available free of charge or as a charged information service with regard to the rules on disclosure control. Learn more about the service: http://www.stat.fi/tup/yritysrekisteri/yhteystiedot_en.html

### Micro-data access

Unit-level data from the statistics on service industry commodities are used for producing statistics and for national accounts calculations and for information service subject to charge.

The data can also be obtained in encrypted form for research use. The use takes place in Statistics Finland's research services. Further information:
http://www.stat.fi/tup/mikroaineistot/index_en.html

### Data revision - policy

Statistics on service industry commodities are not statistics that become revised.

### User needs

Examples of data users outside Statistics Finland are trade associations. In addition, the data are used in compiling the national accounts and the producer price index for services. The data of the statistics are also used at Eurostat.

### User satisfaction

Eurostat assesses the data of the statistics yearly. The last grade received by Finland was very good.

### Quality documentation

A quality report must be submitted to Eurostat within 27 months of the end of the statistical reference year. Eurostat delivers a pre-filled quality report usually in the early part of the year.

### Quality assessment

The quality of the statistics on service industry commodities is examined as the data accumulate. At aggregate level, the data are compared with the previous year and the largest changes are examined.

### Quality assurance

When compiling statistics, Statistics Finland observes the European Statistics Code of Practice (CoP) and the Quality Assurance Framework (QAF) based on them. The Code of Practice concerns the independence and accountability of statistical authorities and the quality of processes and data to be published. The principles are in line with the Fundamental Principles of Official Statistics approved by the United Nations Statistics Division and are supplementary to them. The quality criteria of Official Statistics of Finland are also compatible with the European Statistics Code of Practice. The principles are also compatible with those of the European Foundation for Quality Management (EFQM).

Every year Statistics Finland conducts statistical auditing that helps to ensure the quality of statistics.

The statistics on service industry commodities are official statistics. The Official Statistics of Finland (OSF) are a comprehensive collection of statistics describing the development and state of society. They comprise nearly 300 sets of statistics on 26 different topics. The producers of Official Statistics of Finland have approved a common quality assurance in which they commit to common quality criteria and quality assurance measures. The quality criteria of the Official Statistics of Finland are compatible with the European Statistics Code of Practice. The good practices followed in the statistics are described in more detail in Statistics Finland's Quality Guidelines for Official Statistics handbook.

### User access

The data are released to all users of the statistics simultaneously on the web page of the statistics http://stat.fi/til/palhy/index_en.html. No one has access to the data of the release before the official date of publication. The data are supplied to Eurostat 18 months from the end of the statistical reference year.

## Statistical experts

Riitta Savolainen
Senior Statistician
029 551 3259