Data Structure Definition

Set of structural metadata associated to a Data Set, which includes information about how Concepts are associated with the Measures, Dimensions, and Attributes of a data cube, along with information about the Representation of data and related descriptive metadata.

A DSD defines the structure of an organised collection of data (Data Set) by means of Concepts with specific roles, and their representation.

In order to exchange or disseminate statistical information, an institution needs to specify which statistical concepts are necessary for identifying the series (and for use as Dimensions) and which statistical concepts are to be used as attributes and measures. These definitions form the Data Structure Definition. In a data collection scenario the specification of the Data Structure Definition is often a collaborative venture between the collecting institution and its partners.

There are three types of construct in the DSD: Dimension, Attribute, and Measure. Each of these combines a Concept with its representation (this can be either a reference to a Codelist or a non-coded data type such as "integer", "string", or one of the "date/time" types.

The roles of the three types of construct (Dimension, Attribute, and Measure) are as follows:

A Dimension is an identifying Component, sometimes referred to as a "classificatory variable". When a value is given to each of the Dimensions in a Data Set (this is often called a "Key" or a "series") the resulting Key, when combined with a time value, uniquely identifies an observation. For instance, country, indicator, measurement unit, frequency, and Time Dimensions together identify the cells in a cross-country time series with multiple indicators (e.g. gross domestic product, gross domestic debt) measured in different units (e.g. various currencies, percent changes) and at different frequencies (e.g. annual, quarterly). The cells in such a multi-dimensional table contain the Observation Values.

The DSD construct that specifies the Concept and expected representation of an observation is called a Measure. The semantics of the measure is derived from the Dimensions or a sub set of them and, if not specified in a Dimension, an Attribute indicating the measurement unit e.g. indicator and measure unit (gross domestic product percentage change).

Additional metadata that are useful for understanding or processing the observed value or the context of Data Set or series are called an Attribute in the DSD. Examples of an attribute are a note on the observation, a confidentiality status, or the unit of measure used, or the Title of a series.

A DSD defines the structure of an organised collection of data (Data Set) by means of Concepts with specific roles, and their representation.

In order to exchange or disseminate statistical information, an institution needs to specify which statistical concepts are necessary for identifying the series (and for use as Dimensions) and which statistical concepts are to be used as attributes and measures. These definitions form the Data Structure Definition. In a data collection scenario the specification of the Data Structure Definition is often a collaborative venture between the collecting institution and its partners.

There are three types of construct in the DSD: Dimension, Attribute, and Measure. Each of these combines a Concept with its representation (this can be either a reference to a Codelist or a non-coded data type such as "integer", "string", or one of the "date/time" types.

The roles of the three types of construct (Dimension, Attribute, and Measure) are as follows:

A Dimension is an identifying Component, sometimes referred to as a "classificatory variable". When a value is given to each of the Dimensions in a Data Set (this is often called a "Key" or a "series") the resulting Key, when combined with a time value, uniquely identifies an observation. For instance, country, indicator, measurement unit, frequency, and Time Dimensions together identify the cells in a cross-country time series with multiple indicators (e.g. gross domestic product, gross domestic debt) measured in different units (e.g. various currencies, percent changes) and at different frequencies (e.g. annual, quarterly). The cells in such a multi-dimensional table contain the Observation Values.

The DSD construct that specifies the Concept and expected representation of an observation is called a Measure. The semantics of the measure is derived from the Dimensions or a sub set of them and, if not specified in a Dimension, an Attribute indicating the measurement unit e.g. indicator and measure unit (gross domestic product percentage change).

Additional metadata that are useful for understanding or processing the observed value or the context of Data Set or series are called an Attribute in the DSD. Examples of an attribute are a note on the observation, a confidentiality status, or the unit of measure used, or the Title of a series.

urn:sdmx:org.sdmx.infomodel.conceptscheme.Concept=SDMX:CROSS_DOMAIN_CONCEPTS(2.0).DSD

SDMX, "SDMX Glossary Version 1.0", February 2016

SDMX, Guidelines for SDMX Data Structure Definitions