|Policy name||Metadata policy|
|Date policy was introduced||August 2019|
|This policy has been authorised by (SRO)||Tomas Sanchez Lopez|
|Policy owner||Gareth Watkins|
|Other contacts||Beata Lisowska|
|Scope of the policy||All data in the ONS|
|Next review date||January 2021|
|Release Version||Version 1.0|
Metadata provide information about other data, including a description of the data. This includes information that provides context to the data, for example, how they were collected or the coverage of the data. Metadata can include such information as publication date, description and search keywords. In order to understand the data it holds, the Office for National Statistics (ONS) will provide a comprehensive set of unambiguous metadata relating to the data the ONS manages.
The metadata will be categorised and managed according to the ONS metadata management framework1 , which includes data principles and data standards relating to metadata. All data acquired by the ONS must be accompanied by metadata.
Metadata items will be completed by metadata owners, who will be appointed by the data information asset owners and who will be responsible for periodically reviewing the metadata are correct and of high quality. The metadata owner will initiate corrective action to address any issues found with accuracy and quality of the metadata.
Notes for: Summary
- To be confirmed. The metadata management framework document describing the framework will be published shortly.
Metadata provide information about other data, including a description of the data. This includes information that provides context to the data (for example, how they were collected) or describes the coverage of the data (for example, publication date, description and search keywords).
A metadata strategy describes how an organisation intends to manage its metadata and how it will move from current state to future state practices. The strategy for metadata will evolve over time for the Office for National Statistics (ONS) as the requirements are understood across business domains and the lifecycle of metadata is comprehensively captured across the office through a central metadata management framework. The metadata strategy is part of the ONS data strategy and requires underpinning ONS-wide policies to enforce it. This metadata policy supports the ONS data principles and underpins the metadata management framework, and its adoption is crucial in metadata governance at the ONS.
As part of the metadata management framework1, a business glossary has been created that aims to provide a set of definitions for the main terms used in the ONS.
Metadata is categorised according to the ONS metadata management framework, which has the following categories (and sub-categories):
These are metadata items that identify and define the data and datasets (for example, variables, primary key or identifier, position in the file, data types, valid values, and reference to associated code lists, data models and schemas).
These describe the processing and audit trail information. They ensure the correct handling and management of the data. There are three sub-categories of administrative metadata:
- process metadata, which describe the information related to the processing of the data, including the process used to create the dataset, the start and end times, total records processed, and the input data or parameters used
- paradata, which describe the supplementary monitoring information gathered during the lifecycle of the statistical process, including the time taken to complete the survey or average time taken to complete the survey by respondents or the population and the number of imputations applied to the cleansed or enriched dataset
- technical metadata, which describe the format and structure of the data and datasets as well as the location, number of records and non-statistical elements such as the name of the dataset, the physical location or repository, and the size of the dataset
These describe the data content in a user-friendly manner. They aid user understanding, search and exploration.
These describe the content, methodology and quality of the data including data provenance, source or supplier, general description or purpose of the dataset and its contents, and the frequency or asset owners.
The ONS needs to hold and manage a comprehensive set of unambiguous metadata relating to the data it manages, in order to understand the data, fully extract the value from data and employ appropriate security measures to protect the data.Back to table of contents
This policy covers all the statistical data that the Office for National Statistics (ONS) holds, including those obtained through surveys and from administrative sources in the private or public sectors. The policy is valid for data acquisition and collection by the ONS.
This policy applies to all UK Statistics Authority (UKSA) and ONS employees, including staff on fixed term, temporary or permanent contracts, staff on secondments, students, and contractors.Back to table of contents
The Office for National Statistics (ONS) metadata management framework1 leverages data principles  and data standards relating to metadata.
To improve the usability of metadata, all metadata will be required to comply with the data standard that defines the “minimum set of metadata” for ONS datasets. This means that all data acquired by the ONS from external providers, for example, must be accompanied by metadata that comply with this standard. In practice, captured metadata will go beyond the defined minimum set of metadata.
The metadata must comply with the data principles, in particular:
- data ingestion principles: no data without metadata
- data security principles: legal and regulatory compliance
- data retention and storage: keep data for long enough, do not keep data for too long, and respect local regulations
- common terms and usage: common controlled vocabulary, data processing principles, and maintain change history
5.1 The Information Asset Owner will appoint the Metadata Owner
5.2 Metadata owner
The Metadata owner will be responsible for ensuring the details of the metadata which they own are correct and of high completeness and are managed in accordance with the metadata retention period  they have defined.
The owner will initiate corrective action to address any issues found with the availability, accuracy and completeness of the metadata.
The owner will ensure that metadata is retained to comply with regulations and other existing and future ONS regulatory needs, such as, but not inclusive of, Code of Practice.
5.3 Lead Data Architect
The Lead Data Architect is responsible for informing Metadata owners of any changes (implemented by the Data Architecture team) to this policy and to related data principles, data standards and guidance.
The Lead Data Architect is responsible for ensuring a communication plan is implemented to inform interested parties (e.g. those who are involved in acquiring, creating or amending data and metadata), what their responsibilities are and what processes should be followed to ensure that metadata is managed throughout the lifecycle.
5.4 Data Acquisition / Creation / Amendment
Whoever acquires or creates new data needs to ensure that the corresponding metadata is collected or defined. The metadata must comply with data standards, particularly with the data standards which specify the minimum set of metadata that needs to be provided.
Whoever amends existing data needs to ensure that the corresponding metadata is amended or kept up-to-date and that the metadata owner is informed of the changes.
Whenever new data is acquired / created, a Metadata owner needs to be assigned by the IAO.
Whoever acquires or creates new data or amends existing data needs to check whether the business glossary needs to be updated, and ensure the relevant changes are implemented in a timely fashion.
Chief Data Architect
Stakeholders in Data Architecture division
Head of Knowledge and Information Management
5.6 Glossary term owners
Glossary term owners will either volunteer or be selected to maintain a set of terms in ONS Business Glossary. ONS Business Glossary term owners will assign Subject Matter Experts (SMEs).
The term owners will be responsible to maintain, update, consolidate and link terms to others in the Business Glossary as necessary according to ONS Business Glossary workflows and approval procedures.
6.1 Metadata will be defined and centrally maintained for all data held by ONS from acquisition to dissemination.
Data acquired and collected by ONS from external and internal sources must be accompanied by metadata. The metadata must contain at least a minimum set of metadata elements that are shared across ONS data estate.
Data provided by ONS to external recipient must be accompanied by metadata. This metadata must be up-to-date managed metadata.
New data created by ONS (e.g. statistical output) must be accompanied by metadata.
6.2 The metadata will be categorised and managed according to the ONS Metadata Management framework .
6.3 On the Information Asset Register, information about metadata owners will be recorded
- The Metadata owner will periodically review the quality of the metadata they own and will initiate corrective action to address any issues found with accuracy and quality of the metadata.
- The Metadata owner will assign a retention period for the metadata according to the retention policy .
6.4 Metadata will comprise of a consistent set of minimum metadata elements as defined by the ONS Metadata Model . The minimum set of metadata will include, descriptive, structural and administrative metadata.
- Both the Information Asset Owner and the Metadata Owner will ensure that their assets are described by at least the set of minimum metadata elements.
6.5 The business glossary will be updated with new terms, or the definition of existing terms will be consolidated at the time of data acquisition/ creation or amendment.
- The Glossary term owners will be appointed at the time of data acquisition/ creation or amendment and they will ensure that their terms are up-to-date, linked to other terms, consolidated and deleted if appropriate.
While not applicable at this time, this policy will support the metadata management framework, and the associated procedures will therefore be linked to it.
The procedures for appointing business glossary term owners will follow the same procedures as the ones used by the Knowledge Information Management team in appointing Asset Owners.Back to table of contents
 DAMA DMBOK Data Management Body Of Knowledge, New Jersey: Technics Publications, 2017.
 B. K. Lisowska, C. Bowler and C. Bain, “Metadata Management Discovery Report,” ONS, 2019.
 W. Ram and S. Alder, “ONS Data Principles,” 10 June 2019.
 ONS, Data Architecture Branch, “ONS Data Standards,” 10 June 2019.
 B. K. Lisowska, T. Un Nisa and S. Alder, “Metadata Categories,” 2019.
 C. Bennett, “Data Retention, Data Archiving and Data Destruction Policy”.
 N. Jeenu, “Data Model Document,” February 2019.Back to table of contents