Skip to main content

Research Data Management: Preservation

Beyond the project

Data preservation means more than just making a backup copy of your data. 

Preserving data after a project has ended will:

  • Maintain long-term access 
  • Enable its discovery and usage in the future
  • Prevent data loss
  • Provide support should outcomes of the research be challenged

An active long-term preservation plan should consider:

The preservation of data may also have legal, contractual or ethical considerations attached to it, possibly determined by law, a funding agency or a publisher.

Retention and disposal

Retention outlines the archiving rules as defined by the research funding body or the university e.g.

The minimum recommended period for retention of research data is 7 years from the date of publication. Some data may need to be retained for longer e.g. clinical trials, research involving children etc.

For specific guidance refer to the Curtin Information Management and Archives advice sheets for storing research data.

Click on the image below for a quick reference for retention periods for select types of data. 

Research data retention quick reference

In most cases, not everything associated with the data will need to be retained. Consider:

  • Costs associated with storage
  • Staff effort and time
  • University or funding body requirements
  • Data considered essential to the research project

Data repositories

Repositories are online storage archives that provide a means of managing and facilitating access to digital content, ensuring it remains in an accessible, readable format in the long term. Depositing data in an open access repository may raise the impact of the research as it will be more discoverable and accessible.

These are some examples of repositories:

To deposit data:

Persistent identifiers

Persistent identifiers are labels that uniquely identify online resources (including datasets) by providing a permanent link to them.

Even if your data changes location on the Internet, the identifier remains the same and will still link to the data regardless of the new location.

Two common types of persistent identifiers are DOI (Digital Object Identifier) and Handle e.g.:

  • doi:10.3998/3336451.0003.204
  • hdl:102.100.100/15

Getting a DOI

Curtin University Library can mint DOIs for data using the Australian National Data Service’s Cite My Data service. More information on this service can be found on the Curtin Research Data Collection page.

Data preservation

Digitization at the British Library of a Dunhuang manuscript for the International Dunhuang Project  

More resources


How do I locate data using a persistent identifier?

How do I resolve a persistent identifier?