APSR Workshop – The Data Management Plan: Putting Policy into Practice

On Friday 8 August 2008, I attended the Australian Partnership for Sustainable Repositories (APSR) Workshop, “The Data Management Plan: Putting Policy into Practice” at the University of Melbourne.

Professor Anne Fitzgerald, with whom I work at QUT, gave an excellent and very well received presentation on the legal issues surrounding data management. Her slides can be viewed here.

Here are my notes from the workshop (made roughly during the day):

Data management plans: from idea to reality (10:15am – 10:45am)

Dr Markus Buchhorn (ANU) for Karen Visser

  • We need enduring systems that outlive projects and programs
  • Individuals are human – seven deadly fears:
  1. fear of missed “nuggets” in their data – milk it for everything, for ever and veer
  2. fear of missed errors
  3. fear of unknown custodians/stewards
  4. fear inappropriate leaks (privacy/ethics) – can ruin trust relationships with others
  5. fear the cost of effort
  6. fear lack of recognition
  7. fear trusting someone else’s data

Plan ahead – help researchers to help themselves as far as possible
Build relationships of trust with researchers – engage with researchers as early as possible

Mark Euston (ANU – Information Literacy Program)

  • tasked with developing a training course, workshop and online, for early to mid career researchers, on Data Management Plans (DMP)
  • Objectives of the course –
  1. what is Data Management (DM)?
  2. benefits and requirements
  3. raising awareness of DM services
  4. DMP
  • Manual based on Guidance on Data Management (UK) and Guide to Social Science Data Preparation and Archiving
  • get researchers in by stressing how they can work with their data more effectively and efficiently

What’s happening at… (11:10am – 12:30pm)

Belinda Weaver (UQ)

Issues for the data survey:

  • no ‘joined up’ services
  • no help
  • inequity – not fair – nothing works etc.
  • costs
  • lack of training (people felt insecure about what they were doing)
  • uncertainty
  • no incentive, no rewards

Recommendations from focus group:

  • standardised DM template for funding applications
  • legal advice centralised and accessible
  • service focused support teams for research projects – specific to the discipline
  • survey of all existing data
  • central data storage system
  • develop a clear UQ data management policy
  • templates

Central management of research data – issues:

  • trust
  • data integrity
  • accidental disclosure
  • control
  • sharing
  • re-use (want to know what use has been made of their data – auditing – and if they give data to a person for a particular purpose, they want to know if the person doesn’t end up using the data or not using it for the particular purpose)
  • the long term


  • clear policy and guidelines
  • account manager
  • specialists on teams (want to know who to go to for advice)
  • career path?
  • rewards
  • templates for everything
  • funding to do it properly
  • advice and consultancy
  • institutional support
  • tools (but they want to be told only when they want to be told, and be told how they want to be told)

presentations from workshop available at: http://www.library.uq.edu.au/escholarship/orca.html
UQ developing a expert curation advice service

Lyle Winton (Uni of Melbourne)

  • Uni of Melbourne have a research DMP template
  • looking at training for undergrad students
  • looking at how to keep this up to date
  • possible data management registries
  • from 500 charges of research misconduct, 40% could have been avoided by good data management


Suzanne Clarke (Monash)

  • Monash has a Data Management Committee
  • Research Data Management Toolkit for librarians so they know what to talk about to researchers
  • Identified needs: more education required for researchers on statutory requirements for data, IP and the ownership of research data

Gillian Elliot (University of Otago – NZ)

  • As far as she is aware, NZ has no policies surrounding data management
  • so NZ in quite a different position to Australia
  • Survey in 2007 – researchers in NZ had a lot of data and a lot of stuff loosely stuck together that were unpublished and hard to classify – need help with data management
  • data management and copyright concerned researchers – 48% of survey respondents
  • Atlas of Living Australia; Convention on Biological Diversity; Department of Conservation and Land Information New Zealand; Land Care NZ; National Vegetation Survey Databank

Dr Ashley Buckle (TARDIS – Monash University)

  • TARDIS is a multi-institutional collaborative venture that aims to facilitate the arching and sharing of raw X-ray diffraction images
  • Protein Data Bank – growing exponentially – too much data?
  • Benefits to making raw data available – experiment reproducability/validation

Discussion Groups: Group 2 – Processes for Data Management Planning (1:15pm-2:45pm)

How do we make DM part of the usual research practice?

How can we make raw data count as a citation? – for funding etc. – this is very important, if there is greater recognition of the value data in itself as a citable object then researchers will be more willing to manage their data properly.

Ashley Buckle – we need “data journals” – essentially the same as a database but greater recognition

DM needs to give you a reward at the end that is at the same level as rewards from publication

Better tools – build the researchers tools that are so good that they do not actually realise that they are managing their data.

Reporting back to main group and discussion (2:45pm-4:00pm)

  1. Roles, rights and responsibilities
  • Anne Fitzgerald’s domains of responsibility
  • Policy plus principles
  • disseminate research data as widely as possible
  • develop practical toolkits
  • risk management for universities
  • simple for universities to completing
  • ongoing legal and policy advice
  • insert data management requirements into research proposals and grants
  • get recognition via NHMRC, ARC and ERA to provide regulatory and reward structure
  • need for national centre for legal policy and advice in regard to the data lifecycle including reuse
  • universities to incorporate data management into risk management strategies
  • provide pragmatic family of licences/responsibility statements (like CC) to identify roles and policies
  • DMPs to be built into research project formulation and management

2. Processes for data management planningbetter tools and incentives: build better workflows

  • allowing data management in their modelling: harness tools onto repositories
  • citation: make sure that citation of datasets happens and is rewarded, as incentive for researchers to create good data
  • persuade ARC to make explicit expression of intent in ERA eventually to credit data citation (at least down the road). This as formal submission from this workshop
  • infrastructure: development of a COHERENT NATIONAL NETWORK of repositories, emphasis on discipline specific repositories (though institutionally supported) as a centre for research activity

3. Making it work

  • know what you don’t know
  • each institution needs to:
  1. identify the needs of its researchers (possible role for ANDS here)
  2. map the available services (needs to happen locally)
  3. strategically target the gaps
  4. identify candidate services to drop to fund this
  • Make it easy
  • provide a visible point of contact for the users
  • not necessarily through one channel only
  • not necessarily a one size fits all solution
  • embed regular formal training in how to use services
  • needs to be as easy to use as “MyFlickBook”
  • outreach, marketing, publicity
  • Start small and scale
  1. seed the service and gradual expand it as understanding grows
  2. start with young researchers and use peer group pressure over tie
  3. get good examples going first to generate some quick wins
  4. use growth in tandem with policy
  • Reward innovators in shared services
  1. provide annual performance incentives for going beyond meeting strategic goals
  2. encourage shared services staff to learn new skills
  3. create new job descriptions for new people in management