Forums on institutional responses to the Australian Code for the Responsible Conduct of Research

ANDS has held two information gathering and sharing forums on the Australian Code for the Responsible Conduct of Research to help us develop materials to take to the wider community. The first forum was held in Melbourne on 20th April, 2009 with representatives from Monash University, the University of South Australia, CSIRO, and University of Melbourne.  The second was in Sydney, on 21^st May, 2009, with representatives from the University of Sydney, Australian National University, University of Newcastle, Intersect and Queensland University of Technology. 

These small discussion groups enabled us to compare how institutions are responding to the Code, by developing new policies and procedures. ANDS is seeking to understand the issues involved and provide support in data management.  Common themes emerged over the two days, along with differences in how each institution responded to those themes. 

Strong themes included the importance of support from high levels within the institution, together with a strong governance framework.  It became clear that an institutional response was often collaborative across library, IT, research office, archives and records areas.  The leading role may come from any of those areas, but it is important to have a steering committee or governance structure that facilitates that collaboration.

Issues of access control, storage solutions, training and good practice guidelines on data management and metadata were all discussed, as was the potential for conflict with the agenda of ethics committees.

The importance of high level institutional buy-in was seen to be critical for a coordinated institutional response.  If there is a belief that it will all ‘just go away’, then very little policy is developed, and there are minimal procedures in place.  Where that high level support is present, this paves the way for better structure and support to be put in place.

There was a clear recognition that a critical issue is culture change and good practice, rather than just compliance with a formal code.  Compliance with the Code may be a motivator at executive level, but ease of use is the driver at the user level.  Those who have a strong existing infrastructure are adopting several strategies for driving that change.  There were several key strategies and issues raised.

• One approach is offering training and awareness for higher degree students, helping build good data management skills from the student level up.  Other institutions are using ‘champions’ to help drive change, as they did when implementing research publication repositories. 
• While the Code is seen as a useful motivator, it is only one factor.  A useful strategy would be to try and contextualize the 'data management' issues in section 2 of the Ccode as part of an overall process designed to improve research "integrity". Otherwise it might be seen as another institutionally imposed unecessary add-on and not as an integral part of the research process and institutional shift in the way research is done and accounted for. It needs to be packaged as a change in the external environment that will make people compliant but also more productive.
• The governance framework is important to support this culture change.  Cross section collaboration between library/archives/record management (information management), IT, research office and researchers is critical.  Whoever takes responsibility for data management must ensure good communication between all those areas.  Change does not occur unless you have both the technology and the service.
• There was agreement that policy is relatively easy to develop, and is usually brief.  It is the processes underpinning that policy that are more difficult to put into place. 
• There are many possible models for managing these processes, and there was discussion about interaction between the institutional repository, which is typically used for publications, and research data management.  Some participants are looking at ways of cementing that relationship.  Human effort is also critical - who takes charge of the training and support?  One participant noted that they were adopting the Purdue Model.  (See more information at http://chronicle.com/free/v52/i42/42a03501.htm and http://d2c2.lib.purdue.edu/ )
• Metadata standards for collection description are a new area that needs to be addressed.  What is needed, why is it needed, who creates it and when?  What is the difference between metadata describing an object, and metadata describing a collection?  One participant organisation is currently undertaking a metadata audit, knowing that there are many different metadata schemas for all kinds of objects being used already within their organisation.
• There is value in a data audit process.  Almost all the participants had conducted an institution-wide survey of data management requirements and had found this an extremely useful starting point.  A gap analysis of skills and expertise is also a useful tool.
• Access control is of much interest.  At present there is no spectrum of access control – either material is available to all, or it isn’t.  More granular options are needed for the future.
• Institutions are experimenting with a variety of data management software, including Drupal, Mediaflux, Plone and others.  None has emerged as a clear recommendation for collaboration or data management. 
• Funding is a continuing issue, particularly for storage and digitisation.
• Valuing information was a key topic.  How do you value data?  Data should not simply be stored with no reference to its value.  It requires institutional effort and expertise to value that data. 

As a result of these forums ANDS is preparing a series of half-day seminars to be held around Australia.  One series will focus on the institutional response, and the other on the researcher response.  Details will be released as planning is finalised.

It became clear that there is a desire for a forum to continue discussions on the Code.  ANDS are keen to continue these discussions, so we are currently evaluating appropriate platforms.