The Faculty of Mathematics and Natural Sciences has set up some regulations concerning results published as part of a Master's degree. Here it states (see ยง4.3.4) that "All research results should be available for the University for use in further research." but also that "The student retains the intellectual property rights for his or her thesis. This is regulated in the Norwegian Copyright Act". These regulations cause the necessity for data archiving but in such a way that the students do not loose their property rights.
All students are therefore adviced to archive their data in a Research Data Repository, e.g. the NIRD Research Data Archive, which is a National Data Repository maintained by Uninett Sigma2. When archiving the data in a research data repository, the data is licensed and obtains a PID (a permanent identifier). In addition, the data set can be cross-linked to your master thesis, which also has a permament identifier (a DOI, digital object identfier). In this way, the data can always be tracked to the right person and the right publication. The data is only allowed to be reused according to the licence given to the data set.
Guidelines for data archiving
Not all data produced during your master thesis needs archiving. It's important that the data that is archived is ready to be reused immediately. The data should preferably follow the FAIR principles for data archiving, where 'F' stands for Findable, 'A' stands for Accessible, 'I' stands for Interoperable and 'R' for Reusable. To find out what you should keep or not keep, follow these steps:
- To start, we recommend you to take a look at the Research Data Management Training module by MANTRA, University of Edinburgh. You could also read the UK Data Service 'Practical guidance for anyone working with research data". Both the training module and the practical guide will provide a lot of useful information about your research data and how to archive it.
- If you have acquired your raw data yourself, you need to archive this with good documentation and metadata. With 'documentation' we mean a description of the data set that cannot be described by metadata. This is e.g. a description of how to use the data, the format, etc. The 'metadata' can be entered when archiving the data, and usually contains data like the 'creator name', 'date of creation', 'geographical extent', 'year of creation' etc. The data should be presented in either open formats (e.g. txt, csv, netcdf, etc) or in standard formats (e.g. shp, segy, tiff, jpeg). Raw data that is not acquired by yourself should not be archived. However, you should add a document to the archive describing the raw data set and how it can be accessed.
- A workflow should be presented how the results of the thesis have been obtained. If this is properly described in the thesis itself, this could be mentioned in a short README file that resides with the data set.
- Results should, in the same way as raw data, be presented in open or standard formats. If it is easy to reproduce the results using proper documented in-data and archived code/software, the resulted out-data set can be skipped. Interpretations, which are subject to personal bias (like e.g. seismic interpretations) should always be archived.
- Well documented source code of developed models/software should be archived together with the data set. Eventually, when open source code is used, a document containing information how to download this code should be archived.
- Prepare a rich set of metadata. The metadata is very important for others to find your data.
Data that should NOT be archived
- Your Master thesis (including previous versions). This document is published in DUO.
- Presentations
- Articles (usually pdf-files) collected during your thesis project
- All other documents produced during your thesis period which are not related to your project.
If you have any questions related to this article or archiving research data in general, please contact your local IT support staff (drift@geo.uio.no). Office hours 12.00-13.00 every working day.