Avenida de Manoteras 50-52 - 28050 Madrid
OBJECTIVES:
The Structural Business Statistics: Trade Sector is an annual structural statistical operation aimed at Enterprises engaged in Trade activities (Section G of the CNAE-2009), which includes the sale and repair of motor vehicles and motorcycles, wholesale trade and trade intermediaries and retail trade.
This operation enables the identification of the main structural and economic characteristics of enterprises in the sector studied through a wide range of variables relating to employed staff, turnover and other incomes, purchases and consumption, staff expenditures and investment.
As well as enabling the study of the transformations undergone in the sector studied, the other main uses are as follows:
IMPORTANT NOTE ON THE STATISTICAL CONCEPT OF 'ENTERPRISE':
In accordance with the European Statistical System, the Structural Statistics of Enterprises have implemented for the 2018 financial year a new practical application of the statistical concept of 'Enterprise'. The reasons and details of the adaptation of the statistical concept of the Enterprise were announced by the INE in a Press Release of December 17, 2019.
Under this new approach, a Enterprise can be formed by one or several Legal Units, and in the latter case, the Statistical Enterprise will condense the economic and employment variables of the Legal Units that comprise it. This criterion differs from that previously applied, by which each separate Legal Unit was considered a business. Although the new approach affects only Legal Units that are part of business groups -which are highly relevant entities in terms of economics and employment- the statistical results of the SBS and its distribution by activities and sizes is affected. So that users can compare the SBS data under the traditional approach (based on separate Legal Units) and the new approach (based on Statistical Enterprises), for the 2018 reference year the INE has released both versions of the statistical results.
CLASSIFICATION VARIABLES :
Economic activity. The economic activity carried out by a enterprise is defined as the creation of added value via the production of goods and services. The enterprises often carry out different activities that should be classified by the different classes of the National Classification of Economic Activities. For the purposes of the SBS, enterprises are classified according to their main activity.
Size of the enterprise.The size can be established in terms of the scale of turnover or production value, or by considering the number of people on the enterprise's payroll. In this operation, the second option has been considered to determine the size of the enterprises, establishing size brackets by staff employed.
Geographic distribution, by Autonomous Community. Enterprises can have establishments or premises in different geographical locations and, furthermore, they can also carry out their leadership and administrative management at any of these sites or another located in the same or another Community. The SBS presents a breakdown by Autonomous Community for certain variables (number of premises, turnover, wages and salaries, investment in tangible assets and staff employed) taking the location of these premises into account.
ANALYSIS VARIABLES:
INTEGRATION PROJECT OF THE STRUCTURAL SURVEYS OF THE INDUSTRY, TRADE AND SERVICES SECTORS
Up to and including reference year 2014, the structural business surveys of the industry, trade and services sectors were prepared through separate processes and with different data collection and publication calendars according to the different sectors studied.
From reference year 2015 and with the aim of homogenising and improving the production process of these operations, the statistical results and the release of information, the Integration Project of the Structural Economic Surveys began to be implemented and it has been used over two reference years (2015 and 2016) and has marked an important reform of the aforementioned statistical operations. These statistical operations were renamed, also done in order to highlight the idea of integration, to become the:
Details on the changes and improvements implemented since the 2015 reference year can be found in the document "Structural Business Statistics. Methodology "available on the INE website together with the statistical results of the operation.
NEWS SINCE REFERENCE YEAR 2021 AS A CONSEQUENCE OF THE ENTRY INTO FORCE OF THE EBS REGULATION
The year 2021 was the first reference year in which the new Regulation 2019/2152 of the European Parliament and of the Council on European Business Statistics (EBS Regulation) and its Implementation Act 2020/1197 had to be applied for structural business statistics (SBS). Here are some of its novelties:
To implement these changes, some issues related to the sample design were adjusted, which are described in detail in the Methodology available on the INE website (chapter III and section IV.4)
The detail of the CNAE-2009 activities that constitute this statistics' population scope are as follows:
The basic statistical unit for these operations is the enterprise, which is understood as “the smallest combination of legal units that is an organizational unit producing goods or services, which benefits from a certain degree of autonomy in decision-making, especially for the allocation of its current resources. An enterprise carries out one or more activities at one or more locations. An enterprise may be a sole legal unit “’ (definition of the Regulation of the European Union 696/93).
As previously explained in section 3.1, a new operational concept for 'Enterprise' is applied for the 2018 SES, which we will hereinafter call the Statistical Enterprise and which differs from previous years in that, beginning this year, the Enterprise = Legal Unit analogy will no longer always be true. In other words, some Statistical Enterprises may be made up of two or more Legal Units.
The reporting unit , or rather, the unit from which the basic information is obtained is the Legal Unit. Given that it is perfectly defined and located and has accounting and employment data, the answer is facilitated and homogeneous information is obtained. The Legal Units can be legal persons (mercantile Enterprises) or physical persons (individual entrepreneurs).
Obtaining the basic information from the Legal Units comes either from direct collection (by filling in questionnaires) or, increasingly, from the use of administrative sources (tax data for economic variables, and data from the Social Security for variables related to employment).
And so:
Target population of the Structural Business Statistics: Trade Sector is formed of enterprises, corporations and natural persons whose main activity is described in section G of the National Classification of Economic Activities (CNAE-2009):
From a geographic perspective, this statistic covers the whole of the national territory.
Regarding the publication of results, national results broken down by economic activity are provided, as well as the main results broken down by Autonomous Communities and Cities (5 main variables according to the activity groupings).
The Structural Business Statistics: Trade Sector is an annual statistic.
The data's reference period is, generally, the calendar year. As an exception, for the Legal Units (which are the reporting units) that operate by seasons or campaigns that comprise two separate years and have their data accounted for in this way, their information refers to the season or campaign that ended in the survey's reference year.
There are results available starting from reference year 1999.
There are results available for this statistic starting from reference year 1999.
The implementation of the "Integration Project of the Structural Economic Surveys" throughout the reference years of 2015 and 2016 represented a harmonisation of the dissemination plan for the industry, trade and services sectors, as well as certain methodological improvements. In general, an attempt has been made to maintain the series' continuity and to offer more detailed results. If the changes implemented had an impact on certain variables, the statistical publication included complementary information allow for time comparison. See more details in section 15.2.
For the 2018 reference year, the new practical application of the statistical concept of 'Enterprise' is implemented. This change affects the statistical results of the SBS, and its distribution by activities and sizes. So that users can compare the SBS data under the traditional approach (based on separate Legal Units) and the new approach (based on Statistical Enterprises), for the 2018 reference year the INE has released both versions of the statistical results .
Since the reference year 2019, only the version based on Statistical Enterprises prevails.
The data's reference period is, generally, the calendar year. As an exception, for the Legal Units (which are the reporting units) that operate by seasons or campaigns that comprise two separate years and have their data recorded in this way, their information refers to the season or campaign that ended in the survey's reference year.
Regarding employed staff, to reduce the statistical burden on respondent Legal Units, the SBS questionnaires exclusively request information about staff to 30 September of the reference year. Subsequently, for each Legal Unit, the aforementioned information is combined with the administrative registers on affiliation to Social Security to obtain the value of the average number of employed staff per year.
Data referring to the period: Annual Y: 2022
The compilation and dissemination of the data are governed by the Statistical Law No. 12/1989 "Public Statistical Function" of May 9, 1989, and Law No. 4/1990 of June 29 on “National Budget of State for the year 1990" amended by Law No. 13/1996 "Fiscal, administrative and social measures" of December 30, 1996, makes compulsory all statistics included in the National Statistics Plan. The National Statistical Plan 2009-2012 was approved by the Royal Decree 1663/2008. It contains the statistics that must be developed in the four year period by the State General Administration's services or any other entity dependent on it. All statistics included in the National Statistics Plan are statistics for state purposes and are obligatory. The National Statistics Plan 2021-2024, approved by Royal Decree 1110/2020, of 15 December, is the Plan currently implemented. This statistical operation has governmental purposes, and it is included in the National Statistics Plan 2021-2024. (Statistics of the State Administration).
The Structural Business Statistics: Trade Sector is included in the category of structural surveys according to REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on European business statistics and its corresponding Implementation Act 2020/1197.
https://www.ine.es/normativa/leyes/UE/inditema.htm
The Statistics Institute of Catalonia (IDESCAT) collaborates in the translation of the questionnaires.
To further reduce the statistical burden on respondent Legal Units, the Tax Agency (AEAT), Regional Tax Authorities of the Basque Country and Navarre and the General Social Security Treasury (TGSS) collaborate in this operation, providing administrative files whose information, combined with SBS data for each legal unit, facilitates the reduction of the sample size and the removal of some of the questionnaires' variables.
The Statistical Law No. 12/1989 specifies that the INE cannot publish, or make otherwise available, individual data or statistics that would enable the identification of data for any individual person or entity. Regulation (EC) No 223/2009 on European statistics stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society
In tailored requests the same process is carried out to preserve statistical secrecy.
Up to reference year 2015 inclusive, certain activities up to 4 digits of the CNAE-2009 were catalogued as CETO (Contribution to European Totals Only) when designing the samples for the given sectors. This measure, provided for in article 8.3 of Regulation 295/2008 concerning structural business statistics and applicable only to a limited number of less important activities over the total of each sector, enables the minimisation of increases in sample sizes. This means that the results for these activities are only representative for the preparation of aggregates at the European level and the commitment to not publish the aforementioned data at a national level and, therefore, nor at even a further disagregated level; autonomous or provincial.
The activity codes of the CNAE-2009 to 4 digits considered as CETO until reference year 2015 were:
From reference year 2016 inclusive, thanks to the reform and optimisation of the sample designs, all the activities to 4 digits of the CNAE-2009 in the scope of the study have their corresponding sample support, which made it possible to stop applying the CETO figure. Therefore, from the aforementioned year of publication of Final Results it is possible to offer, for all the activities not affected by confidential data, statistical results up to the level of 4 digits of the CNAE-2009.
The advance release calendar that shows the precise release dates for the coming year is disseminated in the last quarter of each year.
The calendar is disseminated on the INEs Internet website (Publications Calendar)
The data are released simultaneously according to the advance release calendar to all interested parties by issuing the press release. At the same time, the data are posted on the INE's Internet website (www.ine.es/en) almost immediately after the press release is issued. Also some predefined tailor-made requests are sent to registered users. Some users could receive partial information under embargo as it is publicly described in the European Statistics Code of Practice
This statistic is released annually.
The results of the statistical operations are normally disseminated by using press releases that can be accessed via both the corresponding menu and the Press Releases Section in the web
The survey results are released on the INE website (https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176902&menu=ultiDatos&idp=1254735576799) and some results are covered in publications such as the Anuario Estadistico (Statistical Yearbook), Cifras INE (INE Figures), etc.
INEbase is the system the INE uses to store statistical information on the Internet. It contains all the information the INE produces in electronic formats. The primary organisation of the information follows the theme-based classification of the Inventory of Statistical Operations of the State General Administration . The basic unit of INEbase is the statistical operation, defined as the set of activities that lead to obtaining statistical results on a determined sector or subject based on the individually collected data. Also included in the scope of this definition are synthesis preparation.
On INEBase, in the Services section, in Trade, select the Structural Business Statistics: Trade Sector. There, the Results can be explored in the menu on the left.
Access to tables and time series in the INEbase can be found at the following link:
Number of consultations of data tables: AC1=64,365 consultations in 2023.
Number of consultations fo metadata: AC2=1,049.
No microdata file is provided for this statistical operation. Microdata files are only provided to other statistical bodies (Statistical Institutes of Autonomous Communities) that also have an obligation to maintain statistical secrecy.
The data that are released annually aim to offer basic, and relevant, information on main results of the statistical operation, which enables different users' requirements for information to be met.
Results are provided both at a national level, for the different disaggregations by activity, and at the level of Autonomous Communities according to the location of the premises.
It is helpful to note that the year-to-year variations obtained from the results of this survey may be due, on some occasions (to a greater or lesser degree), to changes in the structure of the corresponding reference populations (for example, changes in activity in certain Enterprises that consequently mean that they are classified in a different activity sector to the previous year, etc.), so, not necessarily in all cases, the changes mentioned can be directly associated with a change (increase or decrease), of the same magnitude, in the manufacturing activity of the set of businesses of a certain sector or sub-population.
Finally, it must be highlighted that the release of this statistic is not exclusively limited to the tables offered here or covered in the publication. Safeguarding the restrictions derived from statistical secrecy or from the fact that the survey is sample-based, the existing computing procedures allow for tailored requests for aggregated data to be dealt with, which may be provided in the medium or format chosen by the user.
The request is made through the Information Area:
A detailed description can be found at:
Based on REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on European business statistics, the European Commission (Eurostat) evaluates the quality of the data transmitted and publishes reports on the quality of the European statistics. For this reason, each country provides information on the variation coefficients of the characteristics that are covered in this and at the requested level of disaggregation, as well as data on non-responses. The quality report contains quantitative and qualitative information. Eurostat provides information on the main quantitative indicators, calculated from the data provided by Member States.
https://ec.europa.eu/eurostat/cache/metadata/en/sbs_essbs21_es.htm
This standardised methodological report contains, in sections 10.6 to 17, the elements of what constitutes the "User-oriented quality report" for this operation.
Quality assurance framework for the INE statistics is based on the ESSCoP, the European Statistics Code of Practice made by EUROSTAT. The ESSCoP is made up of 16 principles, gathered in three areas: Institutional Environment, Processes and Products. Each principle is associated with some indicators which make possible to measure it. In order to evaluate quality, EUROSTAT provides different tools: the indicators mentioned above, Self-assessment based on the DESAP model, peer review, user satisfaction surveys and other proceedings for evaluation.
The process of preparing the statistic has established controls to detect and correct errors with the aim of guaranteeing the statistical quality.
The collection of data in Collection Units has programmed for serious and second-tier errors, so data can be filtered as they are entered by the user. Throughout the collection and filtering of data, measures are adopted to reduce non-response.
Finally, once the data arrives at the Central Services, the data are subjected to additional micro- and macro-filtration controls, as well as comparison against other short-term Trade Sector surveys and with information from administrative sources, which enable a final product of sufficient quality to be guaranteed.
Commission Regulation 275/2010 establishes the elements for the evaluation of the data series sent to Eurostat by the different member states for structural business statistics, based on the quality criteria listed in article 12, paragraph 1, of Regulation (EC) No. 223/2009 of the European Parliament and of the Council, namely: relevance, precision, timeliness, punctuality, accessibility, clarity, comparability and coherence.
The statistic can be considered to be of high quality since a sufficiently representative sample is obtained from the target population on a national level and on the Autonomous Community level for each stratum of activity and enterprise size. Furthermore, the data are subjected to adjustment criteria and year-to-year checks, as well as comparisons with administrative sources and other statistical operations.
Users of the statistic include:
The INE has carried out general user satisfaction surveys in 2007, 2010, 2013, 2016 and 2019 and it plans to continue doing so every three years. The purpose of these surveys is to find out what users think about the quality of the information of the INE statistics and the extent to which their needs of information are covered. In addition, additional surveys are carried out in order to acknowledge better other fields such as dissemination of the information, quality of some publications...
On the INE website, in its section Methods and Projects / Quality and Code of Practice / INE quality management / User surveys are available surveys conducted to date.(Click next link)
In the quality section of the INE website, there are different "User Satisfaction Surveys" where the group (Services) in which this statistical operation is classified is evaluated. These results can give an idea of the opinion of users regarding this operation.
The statistic complies with the information requirements according to REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on European business statistics.
It appears in the 2021-2024 National Statistics Plan in the section of structural information for the trade sector and in the EU Statistics Programme: Annual Enterprise Statistics.
Rate of completeness of the data: R1=100%
The sample design intends to reduce sampling errors and the different survey processes are aimed at eliminating or reducing its errors as much as possible, both in the collection stage (rate of response and filtering control) and in the later stages of editing and imputation.
The collection process, coverage control, filtering of errors and imputation of a non-response allows a high level of reliability to be obtained for this statistic.
Sampling errors of the main variables at different levels of disaggregation are calculated.
The tables published include information detailed by activities on the sampling errors of the main variables.
https://www.ine.es/dynt3/inebase/es/index.htm?padre=6638&capsel=6646
Sampling error for Turnover variable: A1=0.06%
Throughout the statistical process a non-sampling error control is carried out. Specific information on the non-response rate is also available.
Over-coverage rate: A2=7.0%
Non-response rate by unit: A4=14.2%
Implementation act 2020/1197 of REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on European business statistics requires the availability of results within 18 months of the end of the reference period. This objective is met by publishing the Final results of the statistics before the end of this deadline.
Opportunity final results: TP2 = t+14 months+ 15 days
The data is disseminated according to the Structural Statistics availability calendar that INE prepares and publishes annually.
The availability of a methodology, a common design and collection, filtering, editing and elevation process throughout the geographic area guarantees the comparability of results between the different Autonomous Communities.
On the other hand, the methodological adaptation to the SBS Regulation makes it possible to compare with the rest of the European Union countries that release the same information for their respective sectors.
For the 2018 reference year, the new practical application of the statistical concept of 'Enterprise' is implemented. This change affects the statistical results of the SBS, and its distribution by activities and sizes. So that users can compare the SBS data under the traditional approach (based on separate Legal Units) and the new approach (based on Statistical Enterprises), for the 2018 reference year the INE released both versions of the statistical results
For the 2021 reference year, there is a break in the series in the 'Number of enterprisess' variable (see information in section 3.1 of this report as well as the details in the Methodology available on the INE website).
Length of the comparable time series under the Statistical Enterprises approach: CC2=5
The use of a single national classification of economic activities allows for the possibility to compare information from the survey with other economic statistics such as, for example, short-term Services and Trade Sector statistics and the Central Business Register (Directorio Central de empresas DIRCE).
It is comparable on an international level, it covers the necessary information required by the various users of the statistic and it is a useful instrument for the National Accounts.
Furthermore, it serves as a framework for the updating of short-term indicators and for studying the transformations undergone in the sector.
Coherence is a fundamental matter both in the planning of the survey's methodology and throughout its preparation. The coherence between variables is compared at all stages of the statistical process.
INE, conscious of the statistical burden on Legal Units, tries to apply certain methods to its questionnaires to reduce this burden as much as possible and, in all cases, it facilitates the completion of information required from respondents.
The following are some of the measures adopted in this operation to reduce the statistical burden:
Intensification of the use of administrative sources. Since the reference year 2012, the EEE had already been using, for a part of its sample (group sample M), methods that allow obtaining the complete questionnaire without resorting to completion by the informant, combining administrative information and imputations based on data from similar units with data ('donors') or values representative of the group to which each unit to be imputed belongs. As of 2018, the proportion of sample units obtained through these techniques has been increasing each year. This strategy has made it possible to considerably increase the total sample size without increasing the number of questionnaires, and therefore the burden on companies. In particular, it has been possible in this way to cover the additional information needs derived from the new practical implementation of the 'Enterprise' concept without increasing the number of questionnaires.
Cooperation arrangements are established to take full advantage of the information available and to avoid unnecessary duplicate statistical operations. The sample size and breadth of the questionnaires are reduced to the bare minimum.
In the sample selection process, negative coordination methods are applied, i.e. minimum possible overlap (covering the survey objectives) between samples of surveys that coexist in the same period of time.
Use of different questionnaire models with different levels of simplification depending on the characteristics of each Legal Unit. Thus, small Legal Units usually fill out a questionnaire with fewer variables and breakdowns. All questionnaire models attempt to reduce the breadth of the questionnaires to the bare minimum.
The change made to the questionnaire for the Integration Project has meant a reduction of certain variables, such as some relating to employment, which will be estimated using administrative data from Social Security.
Internet completion is encouraged by electronic questionnaires with questions, flows and validations adapted to the respondent Legal Unit, which facilitates completion and improves the quality of data collected. Online questionnaire designs are designed in such a way that repeated contact with respondents can be avoided or at least reduced.
The estimate of the budgetary appropriation necessary to finance this statistic, as provided for in the 2024 Annual Programme, is 1,056.75 thousand Euros
The INE of Spain has a policy which regulates the basic aspects of statistical data revision, seeking to ensure process transparency and product quality. This policy is laid out in the document approved by the INE board of directors on 13 March of 2015, which is available on the INE website, in the section "Methods and projects/Quality and Code of Practice/INE’s Quality management/INE’s Revision policy" (link).
This general policy sets the criteria that the different type of revisions should follow: routine revision- it is the case of statistics whose production process includes regular revisions-; more extensive revision- when methodological or basic reference source changes take place-; and exceptional revision- for instance, when an error appears in a published statistic-.
For the 2016 and 2017 reporting periods, the statistical results were published as Advance and Final.
Since the reference year 2018, due to the change implemented as a result of the Statistical enterprise, only the Final results are published.
The Final Results are not subject to review.
SAMPLING FRAMEWORK
The sampling frame is obtained from the Central Directory of Enterprises (DIRCE), which is updated once a year with administrative sources, mainly tax and Social Security, and with information from the statistical operations of the INE.
The DIRCE is an integrated information system at various levels where it stands out, from lowest to highest: establishment, Legal Unit, Statistical Enterprise and business group. For each of these levels, DIRCE contains information on the main economic activity, the number of employees and the turnover, variables that are used in the sample design, and on identification and location data, necessary for a correct collection of the data information.
Since the reference year 2018, the DIRCE includes the new level of the Statistical Enterprise, which is equal to the Legal Unit in the case of independent Legal Units or a set of Legal Units of a business group, or the entire business group (depending on it is determined by the Profiling methodology) for the Legal Units that are part of a business group.
SAMPLE DESIGN AND STRATIFICATION
A distinction is made between the sample design based on the sample of Legal Units (ULE) and that of the sample of Statistical enterprises (EE). The first has been carried out regularly and uses stratified random sampling. In each stratum, a random sample is obtained, except the one formed by LUs with 50 or more employees, in which all form part of the sample. Other relevant units are also included exhaustively in the sample. The sample size is calculated by applying optimal allocation. The estimators are of the Horvitz-Thompson type but adjusted for new registrations (not known at the time of drawing the sample) and, as of the 2021 reference year, calibrated to the 'single number of companies' (a novelty mentioned in section 3.1 of this report).
Indirect sampling is used for the sample of Statistical Enterprises (EE), in the sense that results are obtained by EE from the sample of ULEs. This is based on the methodology detailed by Lavallée and Labelle-Blanchet in their article: "Indirect Sampling applied to Skewed Population", Survey Methodology, June 2013, Vol 39, Statistics Canada.
Details on sample design, determination of exhaustive units, stratification, sample selection, estimators and calculation of sampling errors can be consulted in the document " Structural Business Statistics. Methodology" available on the INE website together with the statistical results of the operation.
The basic information is obtained annually from each Legal Unit, and comes from two sources:
For each reference year t, the collection of questionnaires and the collection of other information from administrative sources is organized as follows:
Completion of the questionnaire over the internet is promoted as much as possible and over 98% of the total questionnaires collected were completed this way. In any case, respondents have other response channels available to them (fax, postal mail or telephone).
The collection units are also responsible for responding to telephone lines to resolve respondents' doubts, and for the recording and filtering of questionnaires. Errors and anomalous data are subject to revision and, if necessary, informants will be contacted again by telephone.
Data validation of the sample Legal Units
During the questionnaire collection stage, the first filtering and coding processes are carried out on them. Both the electronic questionnaires that respondents fill out on the Internet as well as the application of collection management, recording and filtration used by INE's Collection Units have systems programmed to detect errors in order to validate data as they are entered by the user. It distinguishes between serious errors (which must be corrected) and second-tier anomalies (which, after confirmation, must be justified). Furthermore, during the collection and filtering of data, measures are also adopted to reduce non-responses.
The records made by the Collection Units form and feed into, at least fortnightly, the complete recorded files on which the later stages of joint information processing are carried out. These files are processed in the Central Services, where new information coverage control is carried out to guarantee the completeness of the recorded data, to detect duplicates and coverage errors and, at the same time, to carry out an initial assessment of the quality of the variables collected.
A distinction is made between the treatment used for the Legal Units and the additional treatment implemented in 2018 for the Statistical Enterprises.
A. Compilation of the information of the sample Legal Units
As the collection progresses and the complete recording files are formed, the data is subjected to additional micro-filtration controls at Central Services focused, selectively, on the detection and filtering of errors and inconsistencies in the variables of each record, as well as the filtering and imputation of content errors. Depending on the characteristics of each type of error, automatic imputation procedures are used in certain cases. Likewise, systematic errors detected in Previously studies and analyses on recorded data are corrected.
Since reference year 2016 various methodological improvements have been applied to the Structural Business Statistics, included in the ‘Integration Project of the Structural Economic Surveys’. The aforementioned improvements are related to the treatment of Legal Units that cause the target population to rise or fall in the reference year, as well as the imputation of non-response. On the other hand, the estimation of employment variables has been improved by using administrative sources.
Through internal methodological studies, it was concluded that re-weighting non-response led to a certain overestimation of monetary variables, given that a relevant part of non-response is only active during part of the reference year. For this reason, it was decided to replace this re-weighting with the application of different imputation techniques, based on administrative data. The minimum aim of these imputations was that both the turnover and the number of employees correspond to the Legal Units's own value and during the year in question. The rest of the variables are imputed, using highly correlated administrative information, conditional on the two variables, which guarantees the global coherence of the imputed records.
For the effective sample (which from 2016 includes the non-response, as explained above), the elevation factors para Unidades Legales are calculated to determine the estimates of the different variables. The last stage, before the tabulation and release of the results, is obtaining analysis tables through macro-filtration techniques to eliminate the errors and inconsistencies detected. The data are also subjected to comparisons with other statistics (eg, other short-term INE surveys on the sectors studied) and with information from administrative sources, in order to guarantee a final product of sufficient quality.
B. Compilation of information from Statistical Enterprises
For the elaboration of the EEE under the 'Statistical Enterprise' approach, a method has been developed that is based on the following steps, each of which is described in more detail in the EEE Methodology document available on the INE website together to the statistical results of the operation.
The essential idea is that if the Legal Units of a Statistical Enterprise serve, exclusively or mainly, other Legal Units of the same Enterprise (for example, because they sell products under a vertical integration of the production process or provide services as an auxiliary relationship) , said servile Legal Units must be combined with the others to which they support to form the authentic statistical unit "Enterprise", therefore having to combine and consolidate the corresponding variables.
Not applicable as it is not necessary to seasonally adjust the data.