Ethics conversation follow up

As promised here are some links you may find useful.

From ANDS

Ethics, consent and data sharing 

A guide for those engaged in research involving human subjects which is subject to ethics approval, and for those with a role to play in the Human Research ...
ands.org.au/guides/ethics-working-level.html

Sharing Data Ethically

Sharing data ethically: why share data and how to share data.
www.ands.org.au/discovery/ethics.html

National Statement on Ethical Conduct on Human Research  

File Format: PDF/Adobe Acrobat 

Vimeo Webinar

2012-ANDS-Webinars-Ethics-Margaret-Henty 

Those undertaking research involving human subjects are expected to maintain high standards of integrity when it comes to ensuring that ethical standards are met. The issue of how to share data collected in the course of such research is, therefore, something of a challenge. This webinar was designed to set out practical solutions for those who want to share their data and to provide strategies to make it possible. It was based on the recently published ANDS Guide to Ethics, Consent and Data Sharing. Recorded on the 19th April 2012.

How to share data Why share data Research Administrators 

File Format: PDF/Adobe Acrobat
Encourage data sharing by ensuring that ethics forms include relevant questions.
Support institutional initiatives to encourage data management planning ...
www.ands.org.au/discovery/sharing-data-ethically.pdf

Research data policy and the Australian Code for the Responsible ... 

Ethics and privacy — how these affect of the length of data storage and the ability to share. Compliance — what measures are in place to track this. Each of ...
www.ands.org.au/guides/code-awareness.html

Information sources at Australian universities

Finally, as you already know our provide Human ethics, Animal ethics, policies, guides and information. 

Here is a quick link to help you find them using Google search: http://tinyurl.com/ccx7c55

Data Management Plans... DMP21, our new blog

Link: DMP21 blog

As we mentioned at our recent surgeries we hope to adapt the latest version of DMPonline to make it available for you.
Our version will have only 21 baseline data questions, but this can be expanded by you to meet your needs.
We hope this blog will help you understand what we are doing and our rational. We will post our thinking and results so you can follow our progress and ask many questions as we go.
With your help we will make DMP21 available and ready for you to customise and use at your institution,.
As researchers create their data management plans you will be able to use their responses to advise and assist them.

The navigation bar [above] shows the coverage of the questions in DMP21. These are the primary topic areas where researchers will contribute their information. These also become the areas of advice that researchers may require more help and assistance with. We aim to provide sound generic advice and information. DMP21 in addition allows institutions to add their own advice and local links to resources.

Create your own non-resolving persistent identifiers

An early Seeding the Commons project by University of Wollongong needed to create a policy for creating unique keys  for records and persistent identifiers (PIDs) for objects to populate Research Data Australia.
Wollongong did not have access to a handles server or similar infrastructure so this was their answer.

Types of object in RDA

• Parties
• Collections
• Activities
• Services

Each object record requires a ‘key’, one key per RIF-CS record. In addition to this key, any number of ‘identifiers’ can be assigned to an object (within a record). These identifiers may be “persistent identifiers” (such as identifiers issued by the National Library of Australia) and will redirect a user who clicks the identifier, back to a pre-determined webpage.

The following rules are used at UOW as a standard for assigning keys and identifiers to objects in the UOW RDA collection.

Generating Object Keys (one per record)

The Key for all objects will be generated within UOW. They will be of the format:

uow.edu.au/PTY/123 or uow.edu.au/COL/456 or uow.edu.au/ACT/789

where:

uow.edu.au is the unique prefix used by the University of Wollongong

/PTY indicates the identifier is for a party

/COL indicates the identifier is for a collection

/ACT indicates the identifier is for an activity

/SER indicates the identifier is for a service

/### is a unique number

This information provides a good solution for other institutions in the same situation

Generating Object Identifiers (for objects in a record)

Persistent Identifiers are represented in RDA as a clickable link. Ideally for datasets, the link will direct the user to a location where the actual dataset is available for download. However, not all datasets will be available for download. The following principles will be adhered to when entering identifiers for objects into RDA.

• For parties, persistent identifiers will be sought from the National Library of Australia (NLA). Until the NLA identifier is assigned, no identifier will be entered into RDA for that party.
• For collections, a persistent identifier will be manually minted through the ANDS – Identify my Data Service. The guide to this service can be found here
• Where the dataset is available for public download, the ANDS persistent identifier will resolve to the URL at which the data can be downloaded. 
• If the dataset is not available for download, the identifier will resolve to an information page on the dataset or the ANDS Handle Register (i.e. this will be a circular link).
• For activities, no persistent identifier is required for Research Data Australia. The project number assigned by the funding body is used as an identifier. This can be used to search on Google to obtain more information about the project.  A persistent identifier will be minted through the ANDS – Identify my Data Service.

Copyright and Licencing primer

ANDS supports the licensing approach of AusGOAL as it provides a standard and consistent approach which makes it easy for those wanting to re-use data to understand the conditions around data re-use.

ANDS holds periodic webinars on the subject of licensing and the most recent is also available on video. We also offer, through AusGOAL, a consultancy service for those seeking advice on licensing issues. Baden Appleyard, Program Leader of AusGOAL, can be contacted at b.appleyard@ausgoal.gov.au

International versus Australian copyright law

The Australian Copyright Act incorporates the concept of moral rights. It is not possible to give away your moral rights. Under Australian law, it is not possible to extinguish copyright, but it is possible to abandon it.

Who owns IP in data?

Only the owner of the data should assign a licence. In Australian universities there is the issue of who owns the data. ANDS have recently done a review of university IP policies and it is fairly clear that in most universities, copyright in data is held by the university and not by the researcher (unless a student). If the university owns the data, it follows that the university needs a policy about how its data is to be licensed and made available, who makes that decision and what licensing is appropriate.

Can Creative Commons 0 CC0 be used, or is CC BY 3.0 more appropriate?

It is open to question as to whether CC0 is compatible with Australian copyright law. The issue is whether CC0 is prematurely extinguishing copyright or abandoning it. As yet there is no definitive answer to that question, although AusGOAL is arguing that it is the latter. By allocating a CC0 licence, you are giving away the copyright which is bundled with moral rights and this is unlikely to compatible with Australian law. Before advocating the use of CC0, you should check out:

• who owns the copyright in the data?
• who has the authority to apply a licence?
• what kind of licence should be applied?

Using an Auatralian CC BY 3.0 licence does not conflict with our copyright law in this way.

Research Data Management Roadmap

Roadmapping is a methodology that comes from the fields of engineering and technology management. The development of the method is attributed to Motorola over two decades ago. One of his executives stated the following:

"Painting as a Map" by Celia Russell

"roadmaps are extended looks at the future of a chosen field(s) of inquiry; are composed of the collective knowledge and imagination of the drivers of change in a particular field; they communicate visions, stimulate investigations, monitor progress and form an inventory of possibilities of a particular field" (Galvin in 1998 cit. Kostoff and Schaller 2001)

Garcia & Bray 1997 define technology roadmapping as a process for collaborative strategic planning and coordination of ICT developments within organisations. It involves identifying needs, evaluating and selecting appropriate systems as well as planning the implementation. As a result of it, organisations should be better inform and can make better investment decisions.

Ahlqvist et al. 2011 adapt the technology roadmapping method and propose innovation policy roadmapping to address critical innovation policy challenges at national/regional levels within a global context. This method attempts to align the technology development, business models, sociotechnical systems and policy instruments.



Research Data Management Roadmaps need to take into account the technical research data management systems, the university core business objectives, the sociotechnical elements of researchers and their data, as well as research councils policy requirements.

Producing such a roadmap can be a useful exercise for HE institutions embarking in research data management activities to map out their current state and engage in discussions with internal stakeholders to define where they want to be and how to get there. These roadmaps should set targets for institutional infrastructure and services aligned with the institutional strategy and plan the activities and resources to achieve those targets.

But how do institutions make sure that they are setting the right targets?

Higher Education (HE) Institutions in the UK have started creating research data management roadmaps. These roadmaps followed the Engineering and Physical Sciences Research Council (EPSRC) called in April 2012 for UK universities to have a plan to demonstrate how their policies and processes would align with EPSRC's nine research data expectations by May 2015. These expectations provide the framework to help setting the institutional research data management targets:

• The roadmap by the University of Bath uses each of the nine EPSRC's expectations to define their current position and propose a set of activities clearly defining objectives, deadlines and the roles and responsibilities for each of them.
• The University of Surrey Roadmap takes a slightly different approach. They define three phases that start with policy and strategy development, move into service development and policy implementation to finish with training and development. After this they perform a gap analysis against the EPSRC and RCUK data management principles.

In Australia, the research councils have not make a prescriptive call for roadmaps to Australian institutions (at least yet). Nonetheless the Australian Code for the Responsible Conduct of Research's section on research data provides a starting point for a framework to help establishing the targets for  research data management as the code establishes the following responsibilities for institutions:

1. Retain research data and primary materials
2. Provide research data storage and record-keeping facilities
3. Identify ownership of research data and primary materials
4. Ensure security and confidentiality

Monash University has taken an interesting approach in their Research Data Management Strategy and Strategic Plan 2012-2015. After identifying five major research data themes which map to the goals of Monash research strategy they set their vision for 2015 for each of those.

Some of the benefits of doing the research data roadmaps include:

• identify gaps, challenges and opportunities,
• institutions are be better informed for making future investments,
• provide direction to project teams and future activities,
• help building the consensus amongst institutional stakeholders. 

Seeding the Commons projects through their interactions with researchers as well as the institutional service providers are already discovering what is the current state of research data management at their institutions. But is it possible/desired to use these projects to promote this type of process within institutions?

References

Ahlqvist,T., Valovirta, V. Loikkanen, T.  (2011) . Innovation policy roadmapping as a systemic instrument for policy design. Paper presented at Fourth International Seville Conference on Future-Oriented Technology Analysis. May 2011

Garcia, M.L. and Bray, O.H. (1997). Fundamentals of Technology Roadmapping. Strategic Business Development Department Sandia National Laboratories.

Kostoff, R.N. & Schaller, R.R. (2001). Science and technology roadmaps. IEEE Transactions on Engineering Management 48:2, 132–143.

Monash University (2012). Research Data Management Strategy and Strategic Plan 2012-2015. Accessed at:  https://confluence-vre.its.monash.edu.au/display/rdmstrategy/Research+Data+Management+Strategy+and+Strategic+Plan+2012-2015

University of Bath (2012). University of Bath Roadmap for EPSRC. Accessed at: http://www.bath.ac.uk/rdso/University-of-Bath-Roadmap-for-EPSRC.pdf

University of Surrey (2012). Research Data Management Roadmap to 2015. Accessed at: http://t.co/DbSY2Nyz