Reward News Publication
A A A

REWARD 1st Workshop Report

20 December 2011

The first REWARD workshop was successfully run on November 18th. Focusing on the reasons for research data management and best practices, the afternoon event was attended by a balanced mixture of staff and research students.

Anastasia Sakellariadi, our Research Data Scientist, provided an introduction to Research Data Management in archaeology, and gave a brief demonstration of the DMP Online as an example of a tool that can be used to guide researchers through the data management lifecycle.

Brian Hole, the REWARD project manager then gave a brief presentation on why data sharing is important, and which mechanisms are available to make it effective. He covered the fact that sharing data is part of the social contract in academia, and that adhering to this contract using open repositories, data publication and data citation, researchers can receive greater recognition, increased opportunities for collaboration, and enhance their career prospects.

The first case study was then provided by Victoria Yorke-Edwards, who works as a Data Scientist at the MRC Clinical Trials Unit, and lectures in bioarchaeology at Birkbeck College. Vicky looked at the background to data collection in bioarchaeology, and then specifically at the Museum of London’s Wellcome Osteological Research Database (WORD) database, which is designed to both standardise the records of the world’s largest human bone assemblage and to make them easily available for researchers.

This was a very interesting talk that raised a series of issues in discussion. The data that is available is mostly in CSV format, making it accessible to anyone, but the relational database itself can’t be downloaded and queried, making in-depth analysis difficult. One of the biggest challenges is that the database is constantly being updated, leading to the problem of how to cite something without versioning. The website recommends mentioning the year the data was accessed, but without a permanent identifier such as a DOI pointing to the specific version of the data that was referenced, readers will be unable to validate claims made in scholarly publications with any certainty.

The second case study was given by Andrew Bevan, Senior Lecturer at the Institute of Archaeology, who drew on his experience of archaeologists’ sharing behaviour with both spatial and non-spatial data. Andy looked at different kinds of sharing motivations (e.g. communal, authority-based, tit-for-tat, etc.), which led to a discussion of which were most common in archaeology. He also discussed the difficulties arising from the fact that many archaeological datasets have multiple data sources, with different licencing rules. His work on the Greek island of Antikythera provides an excellent example of this, where all of the data can be released openly except for the Greek Army contour map on which it has been plotted. This underlines the fact that any dataset, no matter how small, not archived with an open licence such as CC0, may result in all other datasets into which it is incorporated being either un-releasable or of limited use.

Anastasia then presented the preliminary results of an internal Institute of Archaeology survey on attitudes to data sharing. The results have not yet been fully analysed, but some of the initial statistics are very thought-provoking. When asked whether they had experienced, or expected to experience any problems in sharing their data, participants listed the following factors as most significant:

  • 63%: lack of time to prepare data properly
  • 50%: incompatible data types
  • 41%: lack of technical infrastructure
  • 34%: possible misuse of data
  • 31%: restricted access to data archive
  • 28%: legal issues
  • 25%: lack of financial resources
  • 22%: loss of competitive advantage
  • 6%: no problems foreseen

The most interesting thing about this is that ‘loss of competitive advantage’ is one of the things archaeologists in the survey were least concerned with. This is potentially very good news, as the majority of the other factors can be solved with a combination of training and technical and institutional resources.

The survey also included participants of all age groups and career stages, so it will be interesting to see whether any of the attitudes are correlated with these when further analysis is done. Look out for a future blog post on this in the new year!

Brian then provided an overview of the REWARD project itself, and explained what it is trying to achieve, in tying together all of the previously discussed issues and trying to find a way to incentivise data sharing in archaeology.

The open discussion that followed the presentations was especially interesting. The main topics of discussion were the lack of knowledge of what institutional policies are, and the lack of training available in data management. Nobody attending the workshop knew what UCL’s requirements for open access or data archiving were, which should be a major concern for the university. A recommendation that the REWARD project will make is that the Institute of Archaeology should have a page about this on its own website, with links to the relevant UCL pages. This was seen as important as the institute website is used regularly by staff and students, but finding information in other UCL locations is much more difficult.

It was also felt that there is not enough training in data management built into archaeology courses, and that modules in this should be recommended. The benefit of having a research data scientist in the Institute to provide advice was also recognised, and that this role could be potentially be provided by a TA in future if funding could be made available.

All of the presentations from the workshop are available on the REWARD website resources page: http://www.ucl.ac.uk/reward/resources. We also plan to run the workshop again in February – please contact Anastasia if you are interested in attending.

Page last modified on 20 dec 11 15:42