Turning roadmaps to action

19 March, 2012

Where does a 'roadmap' figure in the transition from policy to practice? This was a key topic last week in in the workshop Developing Institutional Research Data Management Policies in Leeds. I joined DCC staff and more than 40 colleagues from across UK Higher Education there, along with Simon Hodson programme manager of the JISC Managing Research Data programme (MRD02). Most of those present represented projects funded by this programme to develop their institution's support services for managing research data.

The MRD02 projects are tasked with drafting a policy for their institution, a task given a forceful nudge by the EPSRC's deadline of May 1 this year for institutions to produce a 'roadmap', setting out how they will meet this funding body's expectations over the coming 3 years. This deadline also drew people from the DCC's programme of institutional engagement keen to share experiences.

It was clear that the EPSRC mandate has energised university road building towards better data management. It was also apparent that institutions are at different stages in laying out a way ahead, and anticipating effort to persuade some colleagues to get on board. Many have adopted general policy principles like those adopted by University of Edinburgh last year. There was a general view that, while setting out broad expectations and commitments is important, if a policy is to take effect it has to be quickly followed by services offering real value. 

Some participants saw the EPSRC roadmap as an overarching strategic response, both to the policy and to external funders requirements. For others it referred to more action-oriented project plans and documents that, taken together with a policy, demonstrate a commitment to change without corresponding to a single 'roadmap' document.

I don't think these are contradictory views. EPSRC have prompted institutions to develop strategy, which should respond to other funders and legislative changes; just as the overarching policy responds to the RCUK 'Common Principles'. Bearing in mind that the point is to 'align policy and processes' with the expectations it should make sense to structure a strategy document according to relevant institutional roles and  reponsibilities, and the processes or methodology to fulfil them. For example:

  • Policy and strategy development: Draft; Advocate and consult; Ratify
  • Service development/ policy implementation: Initiate change; Assess current provision and practice; Develop or redesign services; Implement and evaluate
  • Training and development: Design and supply training; Evaluate content and delivery

The discussions I was involved in often returned to the question of how to decide what to keep. This is a good example requiring each of the above; appraisal policies that involve researchers and data repositories in informed decision making, and trainng on the kinds of criteria that may be used to value data 

An action plan will be needed to deliver on the strategy. In an earlier post Graham Pryor outlined a tabular format with a column for the timeline and, under each expectation, a listing of:

  • what do we do now and what do we have in place?
  • what must we do to meet any identified gaps?
  • what resources will we commit?

If the action plan is a regularly reviewed working document, a table like this will help monitor compliance. A single table may not be enough; there is overlap between the EPSRC expectations, and across those of other funders. A strategy document can lay out the core messages and high-level requirements along the above lines with a summary of current provision, gaps and resource commitments to be elaborated on in the action plan. In any case, a roadmap will combine aspects of strategy and schedule.

The working documents and resources used to deliver all this will be familiar from other change programmes, such as VLE development. At all levels there will be some combination of advocacy, assessment, diagnosis, redesign, implementation and evaluation of changes made. DCC tools and methods such as DAF and CARDIO can help. Other approaches discussed in this context include scenarios describing common use cases. This can be a helpful first step in defining the workflows that will be required and several workshop participants were developing scenarios as part of their toolkit.

See also....

A compelling view of how RDM responsibilities map on to typical roles found in UK universities is set out by Liz Lyon in her paper The Informatics Transform: Re-Engineering Libraries for the Data Decade

Earlier posts in this series:

Other posts from the Leeds workshop