When you select where to store your research data you should consider the type of data you are working with and match it with a storage solution that meets the various wishes and demands for the data handling.

Are you:

• Generating large amounts of data? 
• Using data that has licensing agreements or other restrictions on it's use and accessibility? 
• Using data that involve human or non-human animal subjects? 
• Working with collaborators outside AAU, that need access to the data?
• Does the data’s data classification match the system's data classification?

When working with active research data, you need to choose a platform or system that meets the data’s requirements and your demands for data handling. So, before deciding what platform, you can use for your data, you need to identify the classification of the data that you are working with. 

• Look at the AAU data classification page, find out what level of data you are working with, and what security demands you need to be aware of.
• When you know what level in the data classification model your data is, then you can find a system that match the security specification in the classification model, your wishes for data handling, sharing, amount of data, and so on.
  Again, have a look at the AAU data classification page. Here you will find the most used systems at AAU matched with the four levels of data in the data classification model, that are safe to store in the system.

There is a risk of data getting lost once a project finishes or the researcher who collected the data leaves the institution. This would mean a big waste of time, money, and knowledge. This is a shame and something we would like to prevent. All research is valuable and should in some way be preserved. 

According to The Danish Code of Conduct for Research Integrity, it is recommended that you preserve the data for a least 5 years. This applies unless otherwise agreed upon.  

Depending on the type of data that you are working with, it is not always possible preserving the data for minimum 5 years. Therefore, be aware if there are any special regulations regarding the preservation of the specific data that you are working with. This is often the case if you are working with health data or data from industry partners – so, always look at the contract and agreement you have made with the data providers.

Considerations when preserving data

The FAIR principles advocate the importance of maximizing the potential of your data and thereby research impact. To do this you need to make your research data Findable, Accessible, Interoperable, and Reusable.

Read more about FAIR.   
 
To make your research data FAIR, start early in the research process to think about what will happen to the data after the research project ends, so you can make good metadata descriptions and associated documentation. FAIR data are supposed to be found,  understood, and reused by others in new contexts.  
 
Choosing, which data should be preserved and ensuring that it is readable and understandable is a challenge but also a rewarding investment. 

In this process, consider: 

• What data do you intend to preserve? Often it doesn’t make sense to preserve all collected and produced data in the research project.
• Data that is valuable to preserve is often data that cannot be reproduced.
• In the case of reused data, it is often sufficient to quote the source.
• All data underlying scientific publications need to be preserved long-term if possible.
• There might be data you need to destroy for legal or ethical reasons. (If not, remember The Danish Code of Conduct for Research Integrity recommendation on 5 years).
   
   
  At AAU you are encouraged to work with the FAIR principles and upload your valuable data in a repository. In a repository, you describe your data with metadata and add a persistent identifier letting others become aware that it exists and what it is. This way, your data can potentially be found and reused in new research. If you are working with data that cannot be public avaliable, you can still be FAIR, if you upload metadata about the data in a repository. In AAU's local repository, DataDeposit you can preserve and upload all types of data.