Skip to Content

Table of Terms with Commonly Used Words

Last modified by Artur K. on 2026/05/29 17:16

Name
 Definition
 Comment
Agency Scheme Maintained collection of maintenance agencies. In SDMX the Agency Scheme contains a non-hierarchic list of maintenance agencies. Each maintenance agency can have a single Agency Scheme, and may have none. The agencies in the Agency Scheme are deemed to be sub agencies of the maintenance agency of the scheme in which they reside. The top-level Agency Scheme is the scheme for which SDMX is the maintenance agency (SDMX Agency Scheme), and every Agency in every Agency Scheme must be related directly or indirectly via intervening Agency Schemes, to an Agency registered in the SDMX Agency Scheme. In this way each Agency can be identified uniquely by the combination of Agencies in the path from the SDMX Agency Scheme to the Agency Scheme in which it resides, plus its own identity in that scheme.
Attachment level Property of an attribute defining the object to which data or metadata are linked. For each attribute specified in a data structure, there is a definition of whether this attribute takes a:
  • value for each observation in the Data Set;
  • value for each time series in the Data Set;
  • value for each group in the Data Set;
  • single value for the entire Data Set.
Some metadata concepts (e.g. frequency) may not be meaningful at the observation level, but only when applied to a higher level (e.g. to a time series of observations). Time, on the other hand, is meaningful at observation level, because every observation is associated with a specific point or period in time. Data Structure Definitions and Metadata Structure Definitions provide information about the level at which a particular concept descriptor is relevant: at observation level, time series level, group level, dataset level or even Agency level. This is known as the "attachment level" of the concept.This is a version 2.0 construct. In version 2.1 this is known as the "Attribute Relationship".
Attribute Relationship Specification of the type of artefact to which a data attribute can be attached in a Data Set. A part of the specification of Attribute in a Data Structure Definition denotes to which part of the data the Attribute can relate in a Data Set. This can be the entire Data Set, specific grouping of the Dimensions, or an Observation.This is a version 2.1 construct. In version 2.0 this was known as the "attachment level".
Codelist Predefined set of terms from which some statistical coded Concepts take their values. The SDMX technical standards are sufficiently generic to allow institutions to adopt and implement any specific representation. However, the use of common Codelists will facilitate users to work even more efficiently as it eases the maintenance of, and reduces the need for, mapping systems and interfaces delivering data and metadata to users. Therefore, a choice over Codelists has a great impact on the efficiency of data sharing.From version 2.1 of the standard it is possible to exchange and disseminate a partial Codelist which is extracted from the full Codelist and which supports the dimension values valid for a particular Data Structure Definition (DSD). The content of the partial Codelist is specified on a Constraint and can be specified for any object to which a Constraint may be attached. This makes it possible to use common (and often quite large) Codelists in multiple DSDs and then to limit their content for use in a specific DSD.
Coherence Adequacy of statistics to be reliably combined in different ways and for various uses. When originating from different sources, and in particular from statistical surveys using different methodology, statistics are often not completely identical, but show differences in results due to different collection methodology concepts, classifications and methodological standards. There are several areas where the assessment of coherence is regularly conducted: between provisional and final statistics, between annual and short-term statistics, between statistics from the same socio-economic domain, and between survey statistics and national accounts.The concept of coherence is closely related to the concept of comparability between statistical domains. Both coherence and comparability refer to a Data Set with respect to another. The difference between the two is that comparability refers to comparisons between statistics based on usually unrelated statistical populations and coherence refers to comparisons between statistics for the same or largely similar populations.In the Data Quality Assessment Framework (DQAF) of the International Monetary Fund (IMF), the term "consistency" is used for indicating "logical and numerical coherence". In that framework, "internal consistency" and "intersectoral and cross-domain consistency" can be mapped to "internal coherence" and "cross-domain coherence" respectively.
Comment Descriptive text which can be attached to data or metadata. In data messages, a comment may be defined as an Attribute and can contain a descriptive text which can be attached to any construct specified in the Attribute Relationship.In Metadata Sets a comment can be attached to any object in the SDMX Information Model that can be identified (known as an "Identifiable Artefact" in the model). For example Agency, Provision Agreement, Dataflow, Code, Concept.In both of these types of messages the relevant Concept (e.g. COMMENT) must be declared in the structure definition (Data Structure Definition or Metadata Structure Definition) together with the object to which it is allowed to be attached in the Data Set or Metadata Set. Note that in a data structure (version 2.1 onwards) it is possible to define the "Attribute Relationship" of any Concept used as an Attribute to more than one of Data Set, group, series, observation. This is not possible using version 2.0. In version 2.0 it is necessary to declare multiple Concepts (e.g. COMMENT_TS, COMMENT_OBS) to achieve this.
Concept Unit of thought created by a unique combination of characteristics. At an abstract level, a Concept is defined in the Generic Statistical Information Model (GSIM) as a "unit of thought differentiated by characteristics". Concepts are used in different ways throughout the statistical lifecycle, and each role of a Concept is described using different information objects (which are subtypes of Concept). A Concept can be used in these situations:(a) As a characteristic. The Concept is used by a Variable to describe the particular characteristic that is to be measured about a Population. For example, to measure the Concept of gender in a population of adults in Spain, the Variable combines this Concept with the Unit Type "person".(b) As a Unit Type or a Population. To describe the set of objects that information is to be obtained about in a statistical survey. For example, the Population of adults in the Netherlands based on the Unit Type of persons.(c) As a Category to further define details about a Concept. For example, Male and Female for the Concept of Gender. Codes can be linked to a Category via a Node (i.e. a Code Item or Classification Item), for use within a Codelist or Statistical Classification.In SDMX the Concept can be given a core Representation such as a reference to a Codelist for an enumerated Representation or other values such as "integer" or "string" for a non-enumerated Representation. This Representation can be overridden in the data structure when the Concept is used as a Dimension or Attribute. A Concept with a core Representation could be regarded as a represented variable.
Contact email address E-mail address of the contact points for the data or metadata.
Contact fax number Fax number of the contact points for the data or metadata.
Contact mail address Postal address of the contact points for the data or metadata.
Contact name Name of the contact points for the data or metadata.
Contact organisation Organisation of the contact point(s) for the data or metadata.
Contact organisation unit Addressable subdivision of an organisation. This contact refers to the contact point for data and metadata.
Contact person function Area of technical responsibility of the contact, such as "methodology", "database management" or "dissemination".
Contact phone number Telephone number of the contact points for the data or metadata.
Data description Metadata element describing the main characteristics of the Data Set in an easily understandable manner, referring to the main data and indicators disseminated. This summary description should provide an immediate understanding of the data to users (also to those who do not have a broader technical knowledge of the Data Set in question).Data can be displayed to users as tables, graphs or maps. According to the United Nations' Fundamental Principles of Official Statistics, the choice of appropriate presentation methods should be made in accordance with professional considerations. Data presentation includes the description of the Data Set disseminated with the main variables covered, the classifications and breakdowns used, the reference area, a summary information on the time period covered and, if applicable, the base period used.
Data Provider Scheme Maintained collection of Data Providers. In SDMX a Data Provider Scheme contains a non-hierarchic list of Data Providers. Each maintenance agency can have a single Data Provider Scheme, and may have none. The identity of the Data Provider is a combination of the identity of the Data Provider Scheme (which includes the maintenance agency) in which it resides and the identity of the Data Provider in that scheme.The Data Provider is the owning organisation of data and reference metadata. These data and reference metadata are reported, exchanged, or disseminated as SDMX Data Sets and SDMX Metadata Sets. The type of data and metadata that are available are specified in a Dataflow and Metadataflow. The union of one Data Provider and one Dataflow or Metadataflow is known as a Provision Agreement.In a data collection scenario the Data Provider is the organisation reporting the data or reference metadata and information can be linked with the Provision Agreement. Information linked to the Provision Agreement can specify where the data or reference metadata are located (data registration) and the data collector (as the Agency of the Provision Agreement) can specify validation Constraints such as allowable dimension values or Series Keys for which data can be reported.In a data dissemination scenario information linked to the Provision Agreement can specify the location of the Data Source and the content of the Data Source in terms of Series Keys available (Constraint).
Data Set Organised collection of data defined by a Data Structure Definition (DSD). Within SDMX, a Data Set can be understood as a collection of similar data, sharing a structure, which extends over a period of time.The Data Set can be represented physically in three fundamental forms:- Generic Data Set: this format allows the representation of data structured according to any Data Structure Definition;- Structure Specific Data Set: this format allows the representation of data structured according to a specific Data Structure Definition;- SDMX-EDI Data Set: a specific case of generic using the UN/EDIFACT syntax and which has limitations on what can be represented. It supports time series only.The Structure Specific format is new to SDMX version 2.1 and combines the functionalities of the version 2.0 Compact and Cross Sectional formats.
Dataflow Structure which describes, categorises and constrains the allowable content of a Data Set that providers will supply for different reference periods. In SDMX, Data Sets are reported or disseminated according to a Dataflow Definition. The Dataflow Definition identifies the Data Structure Definition and may be associated with one or more subject-matter domains. This facilitates the search for data according to organised Category Schemes.A "Dataflow", in this context, is an abstract Concept of the Data Sets, i.e. a structure without any data. While a Data Structure Definition defines Dimensions, Attributes, Measures and associated representation that comprise the valid structure of data and related metadata contained in a Data Set, the Dataflow Definition associates a Data Structure Definition with one or more Category. This gives a system the ability to state which Data Sets are to be reported for a given Category and which Data Sets can be reported using the Data Structure Definition. The Dataflow Definition may also have additional metadata attached, defining qualitative information and Constraints on the use of the Data Structure Definition, in terms of reporting periodicity or specifying the subset of Codes to be used in a Dimension.
Economic sector High-level grouping of economic activities based on the types of goods and services produced.= There is a general agreement on having a high-level breakdown of the economic activity in three main sectors:
  • Primary (extraction, fishing, farming, etc.);
  • Secondary (manufacturing);
  • Tertiary (sales and services).
Some authors add two new categories:
  • Quaternary (information and knowledge-based services);
  • Quinary (human services).
Education level Ordered set which groups and classifies education programmes according to the knowledge, skills, competencies and qualifications which they are designed to impart. The International Standard Classification of Education (ISCED), maintained by the Institute for Statistics (UIS) of the United Nations Educational, Scientific and Cultural Organisation (UNESCO), is used to classify programmes and their resulting qualifications into levels and fields of education. It is a widely-used global reference classification for education systems which provides a comprehensive framework for organising education programmes and qualifications by applying uniform and internationally agreed definitions to facilitate comparisons of education systems across countries.ISCED is the international framework for assembling, compiling and analysing cross-nationally comparable data related to students, teachers, educational attainment and education expenditure. ISCED 2011 is the second major revision of this classification (initially developed in the 1970s and revised in 1997). It was adopted by the UNESCO General Conference in November 2011.
Hierarchical Code Code reference that is part of a hierarchy. The Hierarchical Code references a Code in a Codelist and can have child Hierarchical Codes. It can also reference a Level in a Hierarchical Codelist.
Hierarchical Codelist Organised collection of Codes that may be part of many parent/child relationships with other Codes in the scheme, as defined by one or more hierarchies of the scheme. The Codelist in SDMX can be hierarchical but it is capable of being processed as flat list as each Code can have only one parent code. A Hierarchical Codelist (HCL) is able to have multiple hierarchies and can have formal Levels. The Codes used in an HCL are derived from one or more Codelists therefore an HCL can combine Codes from multiple Codelists and define hierarchies from these Codes. For example, adding geographic codes such as continents or regions.
Hierarchy Classification structure arranged in levels of detail from the broadest to the most detailed level. Each level of the classification is defined in terms of the categories at the next lower level of the classification.In SDMX this is known as a level based hierarchy. SDMX also has the concept of the value based hierarchy where the hierarchy of categories are not organised into formal levels.
isIncluded Construct that indicates whether the contained values of a container object is to be included or excluded from the valid list of values. This is used in validity Constraints to specify if the Constraint lists the items that are included in the list of valid contents, or are to be excluded from the list of valid contents.
Maintainable Artefact Construct that contains structures capable of providing a Maintenance Agency to an object. Maintainable Artefacts inherit the capability of having versioning name, identity and Annotations. In addition a Maintainable Artefact can have an indication that the artefact and its contained items (e.g. the contained items of a Codelist are the Codes) are "final" and there are restrictions on what type of change is allowed without changing the version.
Maintenance agency Organisation or other expert body responsible for the operational maintenance of commonly used metadata artefacts. The maintenance agency is responsible for all administrative and operational issues relating to an artefact or set of artefacts. It is the point of contact for all stakeholders for all issues related to the artefact(s) under its responsibility. The maintenance agency is not a decision-making body. Decisions are made collaboratively among the owners of the artefact.Each identifiable SDMX artefact must have a single maintenance agency (though the maintenance agency could actually consist of several organisations or bodies), either directly (such as Codelist or a Data Structure Definition) or via the container in which it is maintained such as a code (maintained artefact is a Codelist) or a Dimension (maintained artefact is a Data Structure Definition).
Map Correspondence between two or more objects. In SDMX there are several different types of correspondence that are contained in StructureSet artefacts and have different types:
  • StructureMap: Used for mapping Codes in a Codelist;
  • ItemSchemeMap: Used for mapping different schemes such as ConceptSchemeMap, CategorySchemeMap, CodelistMap;
  • HybridCodelistMap: Associates a Codelist and a Hierarchical Codelist.
Each map is a correspondence between the items in one scheme or list and the items in second scheme or list, where the schemes or lists must be of the same type (e.g. Codelists to Codelists).The map can be specified at the level of the Dataflow or Data Structure, or the Metadataflow or Metadata Structure. The Map takes into account the Constraints that are attached to the structural artefact that is mapped.
Member Selection Set of permissible values for one Component of a data or metadata structure. This is a part of a Constraint.
Member Value Single value of the set of values for a Member Selection. This is a part of a Constraint.
Metadata key value Value in a Metadata Set of an identifier Component defined in a Metadata Structure Definition. This is a part of a Constraint.
Non-response error Error that occurs when the survey fails to get a response to one, or possibly all, of the questions. Non-response errors result from a failure to collect complete information on all units in the selected sample. These are known as "unit non-response" and "item non-response".Non-response errors affect survey results in two ways. First, the decrease in sample size or in the amount of information collected in response to a particular question results in larger standard errors. Second, and perhaps more important, a bias is introduced to the extent that non-respondents differ from respondents within a selected sample.Non-response errors are determined by collecting any or all of the following: unit response rate, weighted unit response rate, item response rate, item coverage rate, refusal rate, distribution of reason for non-response, comparison of data across contacts, link to administrative data for non- respondents, estimate of non-response bias.
Relevance Degree to which statistical information meets the real or perceived needs of clients. Relevance is concerned with whether the available information sheds light on the issues that are important to users. Assessing relevance is subjective and depends upon the varying needs of users. The Agency's challenge is to weight and balance the conflicting needs of current and potential users to produce statistics that satisfy the most important needs within given resource constraints. In assessing relevance, one approach is to gauge relevance directly, by polling users about the data. Indirect evidence of relevance may be found by ascertaining where there are processes in place to determine the uses of data and the views of their users or to use the data in-house for research and other analysis. Relevance refers to the processes for monitoring the relevance and practical usefulness of existing statistics in meeting users' needs and how these processes impact the development of statistical programmes.
SDMX Information Model Conceptual model for defining and describing the classes, attributes, and relationships of the SDMX standard. This model is represented in UML (Unified Modelling Language). Section Two of the SDMX technical standard (SDMX Information Model) describes the parts of the model that pertain to structural metadata. Additional structures that relate to Subscription (request to be notified of changes) and Notification (of the changes) are described in Section Five of the SDMX technical standard (Registry Specification).All implementation artefacts such as SDMX-ML and SDMX-EDI specifications for data and structures are derived from the SDMX Information Model and there is a close correlation between the model and these implementation artefacts. This close correlation results in the ability to build syntax and version independent software that can work at the level of the model but which support the various syntaxes and versions of the SDMX implementation artefacts.
SDMX-EDI UN/EDIFACT format for exchange of SDMX-structured data and metadata for time series. SDMX-EDI is a message designed for the exchange of statistical information between organisations in a platform independent manner. The SDMX-EDI format is drawn from the GESMES/TS version 3.0 implementation guide, published as a standard of the SDMX initiative.GESMES (Generic Statistical Message) is a United Nations standard (EDIFACT message) allowing partner institutions to exchange statistical multi-dimensional arrays in a generic but standardised way. GESMES/TS (TS stands for "time series" and the specification is limited to supporting time series data) is an Implementation Guide specifying the use of GESMES for time series data and related metadata, and structural metadata; it can be regarded as a profile of GESMES.In the SDMX standard the GESMES/TS profile is known as SDMX-EDI. It defines the structures of GESMES that are available for use in SDMX-EDI thus allowing partner institutions to design and to build the applications needed to "read" and "write" SDMX-EDI messages.
Statistical classification Set of categories (in the Generic Statistical Information Model sense) which may be assigned to one or more variables registered in statistical surveys or administrative files, and used in the production and dissemination of statistics. The categories at each level of the classification structure must be mutually exclusive and jointly exhaustive of all objects/units in the population of interest. They are defined with reference to one or more characteristics of a particular population of units of observation. A statistical classification may have a flat, linear structure or may be hierarchically structured, such that all categories at lower levels are sub-categories of categories at the next level up.
Statistical subject-matter domain Statistical activity that has common characteristics with respect to concepts and methodologies for data collection, manipulation and transformation. Within SDMX, the list of statistical subject-matter domains (aligned to the Classification of International Statistical Activities maintained by the Conference of European Statisticians of the United Nations Economic Commission for Europe, UNECE) is a standard reference list against which the categorisation schemes of various participants in exchange arrangements can be mapped to facilitate data and metadata exchange. This allows the identification of subject-matter domain groups involved in the development of guidelines and recommendations relevant to one or more statistical domains. Each of these groups could define domain-specific Data Structure Definitions, Concepts, etc.
Statistical unit Entity for which information is sought and for which statistics are ultimately compiled. The statistical unit is the object of a statistical survey and the bearer of statistical characteristics.Statistical units can also be categorised into basic statistical units, i.e. those for which data are collected (also known as observation units), and derived statistical units, i.e. those which are constructed during the statistical production process (also known as analytical units). A basic statistical unit is the most detailed level to which the obtained characteristics can be attached.Statistical units for economic statistics comprise the enterprise, enterprise group, kind-of-activity unit (KAU), local unit, establishment, homogeneous unit of production, etc. In other statistical domains, statistical units can include persons, households, geographical areas, events etc.
Structural metadata Metadata that identify and describe data and reference metadata. Structural metadata are needed to identify, use, and process data matrixes and data cubes, e.g. names of columns or Dimensions of statistical cubes. Structural metadata must be associated with the statistical data and reference metadata, otherwise it becomes impossible to identify, retrieve and navigate the data or reference metadata.In SDMX structural metadata are not limited to describing the structure of data and reference metadata. The structural metadata in SDMX include many of the other constructs to be found in the SDMX Information Model including data discovery, data and metadata Constraints (used for both data validation and data discovery), data and structure mapping, data and metadata reporting, statistical processes.
Structural validation Process to determine the validity of data and reference metadata using structural metadata. In part the validation can be performed by processes that check the syntax of the data for conformance with the standard, e.g. a process for validating an XML instance (e.g. an SDMX Data Set) against the XML schema that defines the allowable structure and content of the instance.In SDMX the structural metadata contain additional metadata that can be used for validation but which cannot be expressed in an XML schema. Examples of these additional metadata include Constraints and Data Providers. The Constraint is used to specify the codes that are contained in a Codelist and which are valid for the type (sub set) of data that are to be expressed in Data Set in given context. The Data Provider specifies which type of data is expected or allowed to be reported or disseminated by a specific individual or organisation.
Structure Set Maintainable collection of Structure Maps that link Components together in a source/target relationship where there is a semantic equivalence between the source and the target Components. The Structure Set can contain maps between two Item Schemes of the same type: Codelist, Concept Scheme, Organisation Unit Scheme, Data Provider Scheme, Data Consumer Scheme. The Structure Set can also contain a map between two Data Structures i.e. map of the Dimensions and Attributes and corresponding code values where these are also mapped.A typical use of Structure Sets is to provide mappings between an SDMX data structure used in an internal system with an SDMX structure of an external dataset when imported to or exported from the internal system.
Time format Technical format for the representation of time. The technical time format and its related Codelist are part of the technical standards for SDMX-EDI and SDMX-XML.In version 2.0 of SDMX there is a recommendation to use the time format attribute to gives additional information on the way time is represented in the message. Following an appraisal of its usefulness this is no longer required. However, it is still possible, if required, to include the time format attribute in SDMX-ML.
Title Textual label used to refer to a statistical object. "Title" may be used as a semantic name describing a statistical object.In SDMX, a title can be referred, for example, to a time series as a "time series title", or to an Observation as an "observation title". This Concept may be used several times in a Data Structure Definition (DSD) by suffixing the ID corresponding to the attachment level, e.g. TITLE_TS (series level), or TITLE_OBS (observation level).
Title complement Detailed textual label used to refer to a statistical object. "Title complement" is an additional name to "Title" describing a statistical object.In SDMX, a title complement can be referred, for example, to a time series as a "time series title complement", or to an Observation as an "observation title complement". This concept may be used several times in a DSD by suffixing the ID corresponding to the attachment level, e.g. TITLE_COMPL_TS (series level), or TITLE_COMPL_OBS (observation level).
Unit of measure Unit in which the data values are expressed. The unit of measure is a quantity or increment by which something is counted or described, such as kg, mm, °C, °F, monetary units such as Euro or US dollar, simple number counts or index numbers. The unit of measure in connection with the unit multiplier, provides the level of detail for the value of the variable.For data messages, the concept is usually represented by codes. For metadata messages the concept is usually represented by free text.
Versionable Artefact Construct that contains structures capable of providing a version to an object. The version is mandatory and other attributes (such as "to" and "from" validity dates) are optional. Versionable Artefacts inherit the capability of having names, identity and Annotations.

© Semantic R&D Group, 2026