Data Management Plans: Outline and Examples
Other Contributers and Roles:
Data Types and Structure
Describe the data types and structures here. For instance data might be tabular in CSV text format or data might be a series of PNG image files representing a time series, each time series in its own directory.
You should also describe any metadata you will be collecting. This can be instrument settings, image capture methods etc. There are a large number of metadata standard. Feel free to contact us for help.
Data Acquisition, Integrity and Quality
Describe how you will acquire your data and ensure that it is not lost or corrupted. Some common methods are to use incremental back up of data, md5 hash algorithms to protect against corruption, storage on raid arrays to protect against hardware failure.
Privacy and Sensitive Data Issues
Address any privacy issues here. If you have data that contains personal information, or data of a secret nature, you need to describe how you will protect your data and prevent others from gaining access to the personal or secret information.
State what rights you wish to maintain over your data. Consider if you wish to be attributed (cited), if you wish to allow commercial use of your data and if you allow your data to be modified. You can also embargo your data for a period of time to give yourself time to publish from your data. A creative commons license is a great place to go.
Also, if you are using someone else's data, you need to handle that here.
How will you make your data available. The previously common "Available Upon Request" is likely not to be sufficient. If you have a disciplinary archive, you may submit your data there. Also some people make their data available through the journals where they publish or on personal web sites. We are also working to make LoboVault, our institutional repository, a destination for your data. The text below describes this scenario:
The data will be archived in perpetuity at the University of New Mexico repository. The data will be available [upon creation | upon conclusion of the grant | after some embargo period] in accordance to the rights policies outlined above.
The data archive at the University of New Mexico is still being developed, but is being built using the Open Archive Information Systems (OAIS) model as its organizing paradigm. The DSpace repository uses Qualified Dublin Core for descriptive metadata. The data archive will keep a separate metadata record in XFDU. This XFDU record will use PREMIS as the primary administrative metadata schema. Additional technical metadata schemas will be incorporated into the record in accordance to current standards in the field. The PI will be responsible for retaining the required metadata, the Data Librarians will organize and format the metadata, and work with the PI to ensure its completeness and accuracy.
Data will be archived in LoboVault, the University of New Mexico’s DSpace repository. Primary responsibility for curating and preparing the data for archiving rests on the Data Librarians at the University of New Mexico Libraries.
Example Data Management Plans