Research Data Management: Data Lifecycle

This diagram has been created by the team of data steward at the University of Bologna (Alma Mater Studiorum - Università di Bologna, UniBo) in October 2022. It proposes a data lifecycle model inspired by the University of Virginia Library’s model (https://guides.lib.virginia.edu/c.php?g=515290&p=3522215). It has been developed in parallel to the Research Data Management Decision Tree, available here: https://doi.org/10.5281/zenodo.7190004

Emphasis is put on a careful planning of data management, which should always precede data collection (and re-use of existing data). A reference to the opportunity of creating and maintaining a Data Management Plan (DMP) has been added. This is not always compulsory, but is increasingly required by funders.

Collecting, analysing and storing data (and possibly sharing them with a group) remain at the heart of the lifecycle, constituting what we called “data handling”. Here, the process is not linear, and researchers tend to move from one stage to another in a recursive fashion.  

At any point during data handling, it is possible to deposit data: the responsibility for storing and safekeeping is passed on to the repository, the time-scale shifts from short-term to long-term, and data become citable (and possibly discoverable) by the wider scholarly community and beyond. Importantly, deposited data can always be re-used as the basis for a new round of collection/analysis/storage that will lead to a new deposit, and so on.