Prepare your data collection for deposit

Preparation for deposit should take place throughout the data lifecycle, starting from the data collection. Especially for research involving human participants, decisions about participant communication, anonymisation and documentation directly affect whether data can be accepted for reuse and how it can be shared.

Data deposited with the UK Data Service must represent the final, cleaned, quality-assured and documented version of the data, suitable for secondary analysis. All planned data cleaning, validation and anonymisation procedures, alongside documentation processes, must be completed before submission.

All data deposits require comprehensive collection-level metadata to support appraisal, processing, discovery and reuse.

Depositors should be prepared to provide information covering:

Study identification (e.g. descriptive title, alternative title).
Description of the data collection (abstract, aims, topics covered).
People and organisations involved in the creation of the collection (data creators, contributors, rights holders, contact details).
Funding and project information (where applicable).
Methodology (including population and geographical coverage, temporal coverage, data collection methods, sampling, observation units, type and kind of data).
Ethics, consent, disclosure risk and anonymisation (where applicable) *.
Access conditions and re-use considerations.
Description of data and documentation files (formats, types, sizes).

The exact metadata fields and when they are completed depend on the repository where the data are hosted (e.g. in-house curated or self-curated ReShare). However, all required metadata must be completed before a deposit can be finalised.

Failure to provide this information will delay the deposit process.

*For research involving human participants, depositors must provide copies of the Participant Information Sheet(s), consent form template(s) and privacy notice(s) used in the study. Where such documentation does not exist, depositors must provide information on the legal and ethical basis for data collection and use, including any data sharing agreements or governance approvals.

Whether depositing large-scale population-representative collections in the UK Data Service in-house curated repository or smaller and medium-sized data collections via the self-curated repository, ReShare, data creators should follow the guidance below when preparing data collections.

As part of our appraisal and deposit review process, we assess whether appropriate steps have been taken to prepare data for safe and sustainable reuse. Following this guidance will help ensure that collections meet ethical and legal requirements and can be made available to researchers.

In addition to the summary points noted below, we provide comprehensive best practice data management guidance for data producers, their teams and research support staff, which can be found in the Research data management section of our website.

We run a regular training workshop programme covering key areas of managing and sharing research data. Please contact us if you would like to discuss any of these issues further.

When preparing data collections that include information about people, data producers must address privacy protection as part of the deposit process. This should be considered during data collection and before the deposit process.

Privacy protection for data collected about people must be addressed through a combined approach:

Transparency through documented communication practices with the participants.
Technical protection through pseudonymisation and anonymisation.
Governance through access controls and usage conditions.

These measures work together and must be considered in combination. Consent and transparency alone are not sufficient to manage disclosure risk, and anonymisation does not remove the need for appropriate access controls.

Data privacy

If your data include information about identifiable individuals, contain potentially disclosive information or are derived from administrative or confidential sources, ensure that:

You have documented the lawful basis and any applicable acts for data processing and data sharing.
Transparency requirements (e.g. participant information) have been addressed.
Appropriate anonymisation or pseudonymisation has been applied where necessary.
A disclosure risk assessment has been undertaken.
Appropriate access conditions (Open/Safeguarded/Controlled) have been selected.

If your data are fully aggregated, non-human or already public, historical and non-sensitive, anonymisation may not be required. Please document why privacy protections are not applicable.

Copyright and intellectual property

If your data include third-party materials or content created by others, ensure that:

Copyright or licensing restrictions have been reviewed.
Permission has been obtained where required (e.g. survey instruments, scales, proprietary classifications, linked datasets).
Any reuse restrictions are clearly documented in the deposit including any attribution requirements specified where applicable.

For data that incorporate variables derived from secondary data sources, linked datasets, or external resources, depositors must complete and submit a Variable Information Log (Excel). The purpose of the log is to ensure transparency regarding the origin and licensing of all externally sourced variables included in the deposited data.

Before submitting, ensure that:

The dataset represents the final, cleaned and quality-assured version suitable for reuse.
All planned validation and consistency checks have been completed.
Temporary, draft, or administrative files/variables have been removed.
All required documentation and metadata needed to interpret and reuse the data are included in the deposit.
Filenames are meaningful and consistent and do not contain special characters. Avoid spaces and capital letters; use underscores (_) between words and keep filenames concise to avoid file path length issues when unzipping.
Recommended formats are used, and no information was lost during file format conversion.
Individual files are not password-protected. Where encryption is required for transfer, use a password-protected ZIP archive only; the original files within must remain unprotected to preserve file integrity and metadata.

If any of the above are not complete, further preparation is required before submission.

Ensure that a study-level document (e.g. user guide or technical report) is provided. This must describe:

The purpose, context and design of the study.
How the data were collected, generated or compiled (including sampling, selection criteria or data sources where applicable).
Data processing and preparation procedures.
Any anonymisation, transformations or derivation.
Quality assurance procedures applied to the data.

Supporting materials used to collect or generate the data (e.g. questionnaires, interview guides, instruments or protocols) must also be provided where applicable. These may form part of the main study-level document or be submitted as separate files.

In addition, for collections containing multiple files or complex structures, a ReadMe file providing an overview of the collection must be submitted; a template ReadMe file (Word) is available.

Tabular data collections (e.g. survey data, administrative records, experimental data)

Ensure that:

Variable names are meaningful, consistent and kept short; avoid spaces and special characters.
Variable and value labels are complete and accurate.
Missing value codes are defined and documented.
Units of measurement are specified where applicable.
Redundant, temporary, or administrative variables have been removed.
A codebook or data dictionary is included. At minimum it must include Dataset Name; Dataset Description; Variable Name; Variable Description/Label; Data Type; Values (where applicable) and Length (where applicable).
Free text or string variables have been reviewed for direct identifiers and unintended sensitive content.

If the data are longitudinal, ensure that:

Wave identifiers are included.
Changes in variables, methodology, or sampling across waves are documented.

Text data collections (e.g. transcripts, fieldnotes, documents)

Ensure that:

Consistent formatting has been applied, including header metadata in each file.
Transcription conventions used are described (e.g. notation for pauses, emphasis, non-verbal cues) where applicable.
Personal identifiers have been removed or replaced where necessary.
A structured data list (Excel) is provided for interview-based collections, summarising key metadata for each interview (e.g. interview ID, contextual information, topics covered, file names).

See our recommended model transcription format (PDF) and example transcription instructions (PDF) for guidance on preparing qualitative transcripts.

Image, audio and video data collections (e.g. photographs, recorded interviews, observational video)

Ensure that:

Technical metadata (e.g. format, duration, resolution, file size) are included in the deposit, either within the study-level documentation or in a separate metadata file.
The recording context (e.g. date, location, setting and purpose of recording) is described.
Transcripts or captions are included where available.

Spatial data collections (e.g. shapefiles, raster grids, mapped boundaries)

Ensure that:

The Coordinate Reference System (CRS) is specified.
Variables or attribute fields are described (e.g. via a codebook or metadata file).
Spatial vector files have been checked to ensure geometries are valid and free from errors.
Any spatial masking, aggregation, jittering or boundary modification applied to reduce disclosure risk is described.

Relational (network) data collections (e.g. social networks, interaction graphs)

Ensure that:

A node list and edge list (or equivalent relational tables) are included in the deposit.
Unique identifiers are clearly defined.
The network structure is specified (e.g. directed/undirected, weighted/unweighted).
The rules used to construct relationships are described (including data source and inclusion criteria).
Disclosure risk has been assessed and appropriate access conditions selected where necessary.

In some circumstances, a metadata-only record may be created in ReShare. This may apply where:

Data cannot be shared due to legal, ethical or contractual restrictions (subject to prior discussion and, where required, formal approval or waiver), or
Data have been archived in another trusted repository, and a UK Data Service catalogue record is created to support discoverability of social science data.

For ESRC-funded research, engagement with the UK Data Service is required even where data are not deposited. Where applicable, a metadata record must be created describing the project and data collection(s). Metadata records should be completed in as much detail as possible and include any documentation that can be shared (e.g. questionnaires, instruments, topic guides). Where data are archived in an alternative repository, the ReShare record must include clear links to the external archive location (e.g. DOI or persistent URL) under Related Resources.

This site uses necessary cookies

Website stats

Prepare your data collection for deposit

Data privacy

Copyright and intellectual property

Tabular data collections (e.g. survey data, administrative records, experimental data)

Text data collections (e.g. transcripts, fieldnotes, documents)

Image, audio and video data collections (e.g. photographs, recorded interviews, observational video)

Spatial data collections (e.g. shapefiles, raster grids, mapped boundaries)

Relational (network) data collections (e.g. social networks, interaction graphs)

More help in 'How to deposit data'

This site uses necessary cookies

Website stats

Prepare your data collection for deposit

Collection-level metadata

Guidance on preparing and managing data

Sharing data about people

Ethical and legal considerations

Data privacy

Copyright and intellectual property

Deposit readiness checks (all deposits)

Documentation requirements (all deposits)

Data-specific requirements

Tabular data collections (e.g. survey data, administrative records, experimental data)

Text data collections (e.g. transcripts, fieldnotes, documents)

Image, audio and video data collections (e.g. photographs, recorded interviews, observational video)

Spatial data collections (e.g. shapefiles, raster grids, mapped boundaries)

Relational (network) data collections (e.g. social networks, interaction graphs)

Catalogue-only records (metadata records)

More help in 'How to deposit data'