There is a wide and growing interest in promoting Research Data Management (RDM) and Research Data Sharing (RDS) from many stakeholders in the research enterprise. Funders are under pressure from activists, from government, and from the wider public agenda towards greater transparency and access to encourage, require, and deliver improved data practices from the researchers they fund.   Funders are responding to this, and to their own interest in improved practice, by developing and implementing policies on RDM and RDS. In this review we examine the state of funder policies, the process of implementation and available guidance to identify the challenges and opportunities for funders in developing policy and delivering on the aspirations for improved community practice, greater transparency and engagement, and enhanced impact.   The review is divided into three parts. The first two components are based on desk research: a survey of existing policy statements drawing in part on existing surveys and a brief review of available guidance on policy development for funders. The third part addresses the experience of policy implementation through interviews with funders, policy developers, and infrastructure providers.   In our review we identify, in common with other surveys, that RDM and RDS policies are increasingly common. The most developed are found amongst funders in the United States, United Kingdom, Australia, and European Union. However many other funders and nations have aspirational statements or are developing policy. There is a broad pattern of policy development moving from aspiration, to recommendations, to requirements, and finally reporting and auditing of data management practice.   There are strong similarities across policies: a requirement for data management planning, often in grant submissions, expectations that data supporting published articles will be made available, and in many cases requirements for data archiving and availability over extended periods beyond grants. However there are also important differences in implementation.   There is essentially no information available on the uptake and success of different policies in terms of compliance rates, or degrees of data availability. Many policies require a Data Management Plan as part of grant submission. This requirement can be enforced but there is disagreement on the value of this. One view is that requirements such as DMPs are the only way to force researchers to pay attention to these issues. The other is that such requirements lead to a culture of compliance in which the minimal effort is made and planning is seen as a “tick-box” exercise that has no further value. In this view requirements such as DMPs may actually be damaging the effort to effect culture change towards improved community practice.   One way to bring these two views together is to see DMPs as living documents that form the basis of collaboration between researchers, funders, and data managers throughout the life of a research project. This approach is reflected in guidance on policy development that emphasises the importance of clarifying responsibilities of various stakeholders and ensuring that researchers are both recognised for good practice and see tangible benefits.   More broadly this points to the need for the program of improving RDM and RDS to be shared project with the incentives for funders and researchers aligned as far as is possible. In the interviews successful policy implementation was often seen to be dependent on funders providing the required support, both in the form of infrastructure and resourcing, and via the provision of internal expertise amongst program managers. Where resources are limited, leveraging other support, especially from institutional sources, was seen as important as was ensuring the scope of policy requirements were commensurate with the support available and readiness of research communities.   Throughout the desk research and the interviews a consistent theme is the desire for cultural change, where data management and sharing practices are embedded within the norms of behaviour for research communities. There is general agreement that progress from aspirational policies to achieving compliance is challenging and that broad cultural change, with the exception of specific communities, is a long way off. It is interesting to note that discussion of cultural change is largely externalised. There is little engagement with the concept of culture as an issue to consider or work with and very little engagement with models of how cultural change could be enabled. The disagreement over the value of DMPs is one example of how a lack of active engagement with culture and how it changes is leading to problems.        Key Findings      Policies on RDM and RDS are being developed by a number of agencies, primarily in the Global North. These policies are broadly consistent in aspiration and outlines but differ significantly in details of implementation.   Policies generally develop along a path starting with aspirational statements, followed by recommendations, then requirements, and finally auditing and compliance measures.   Measurement of policy adoption and compliance in terms of the over goals of increased availability and re-use of data is not tracked and is likely unmeasurable currently.   Data Management Plans are a central requirement for many policies, in part because they can be made compulsory and act as a general focus for raising awareness.   There are significant differences in the views of stakeholders on the value of Data Management Planning in its current form.   Some stakeholders regard them as successful in raising awareness albeit with some limitations.   Some regard them as actively damaging progress towards real change in practice by making RDM appear as one administrative activity among the many required for grant submission   Successful policy implementation is coupled with funder support for infrastructure and training. Seeing RDM as an area for collaboration between funders and researchers may be valuable   Internal expertise and support within a funder is often a gap which becomes a problem with monitoring and implementation   DMPs can be a helpful part of process but it will be important to make them useful documents throughout and beyond the project   If the object of RDM and RDS policy is cultural change in research communities then direct engagement with understanding the various cultures of researcher and other stakeholder communities, alongside frameworks of how they change is an important area for future focus.

In this talk, I use Mauss' theory of the gift in order to shift how we think about data sharing and data donation, particularly in the context of biomedical res

Journal Impact Factors are unreliable and may be gamed. But can they still offer value?

Objective The purpose of this study is to describe the volume, topics, and methodological nature of the existing research literature on research data management in academic institutions.   Materials and methods We conducted a scoping review by searching forty literature databases encompassing a broad range of disciplines from inception to April 2016. We included all study types and data extracted on study design, discipline, data collection tools, and phase of the research data lifecycle.   Results We included 301 articles plus 10 companion reports after screening 13,002 titles and abstracts and 654 full-text articles. Most articles (85%) were published from 2010 onwards and conducted within the sciences (86%). More than three-quarters of the articles (78%) reported methods that included interviews, cross-sectional, or case studies. Most articles (68%) included the Giving Access to Data phase of the UK Data Archive Research Data Lifecycle that examines activities such as sharing data. When studies were grouped into five dominant groupings (Stakeholder, Data, Library, Tool/Device, and Publication), data quality emerged as an integral element.   Conclusion Most studies relied on self-reports (interviews, surveys) or accounts from an observer (case studies) and we found few studies that collected empirical evidence on activities amongst data producers, particularly those examining the impact of research data management interventions. As well, fewer studies examined research data management at the early phases of research projects. The quality of all research outputs needs attention, from the application of best practices in research data management studies, to data producers depositing data in repositories for long-term use.

Interesting theory-of-change case study?

Predatory journals – which charge high fees and often offer little-to-no vetting of research quality – are a problem, and lately an easy target for authors eager to spoof the problems of the publis…

Heuristics beat judgment. EVERY DAMN TIME.

To the Rescue of the Orphans of Scholarly Communication presentation at CNI Spring 2017 meeting Herbert Van de Sompel http://orcid.org/0000-0002-0715-6126 Mich…

For many years, Jeffrey Beall single-handedly fought his own (often controversial) battle against ‘predatory’ publishers, mostly via his ‘Scholarly Open Access’ blog (AKA Beall’s List) – a battle he ultimately seemed to have lost when the blog was taken offline in mysterious circumstances in January.

Authors of scientific papers often don’t want to share their data, sometimes claiming that the data are proprietary. But there are solutions to that.

From how we do science to publishing practices to the sociology of science, there isn’t an aspect of the scientific endeavor that isn’t in flux right now. Long-term readers of Jabberwocky know that…

"Needless to say, these three scenarios are fundamentally incompatible. Yet more incompatible scenarios can be generated from the perspectives of funding agencies, university administrators, private business, students, and other stakeholders. Specifics of the scenarios will vary greatly by research specialty, institution, career stage, funding source, and other factors."

Pressure to publish short articles removes details, leaves readers confused.