of the central questions of research policy is how to organise and
carry out research in order to maximise the benefits that derive from
it. This question prompts heated debate. Should we prioritise basic
research over applied research or vice versa? Which disciplines
are most impactful, and so deserving of most funding? Are there
particular disciplines that should be favoured in order to align with
national priorities for industrial strategy, or societal challenges?
Important new insights on these questions are being revealed from the
new dataset that is available following the Research Excellence Framework (REF); nearly 7000 case studies of impact from research are now available
for mining and analysis. While the dataset has some limitations, the
case studies still present an unprecedented opportunity to start to
answer these questions.
Some insights are already available thanks to the high level analysis carried out by the Policy Institute with the Department of Digital Humanities at King’s College London, working in partnership with Digital Science to
create an online searchable database. One of the most striking points
to emerge so far is the relationship between research disciplines and
impact. Using semantic analysis the King’s/Digital Science team have
shown that a very large proportion of the case studies draw on
underpinning research from diverse disciplines. While this has been
proposed before, especially in the context of impact on societal
challenges, and is supported by smaller scale studies, the analysis of the case studies is the first large scale systematic analysis to support the conclusion.
This is a finding of some significance. Impact comes not from
research in specific disciplines, but from the integration of knowledge
and understandings from a range of disciplinary inputs. The interactions between disciplines matter.
A focus on integration of different disciplines raises some questions
about how and where that integration takes place. There are a number of
levels where knowledge can be integrated. Knowledge can be brought
together for the delivery of impact, or there can be varied levels of
integration within the research process itself. On the one hand
integration can be a component of knowledge exchange, on the other
impact can build on interdisciplinary research. The King’s/Digital
Science analysis doesn’t distinguish between these different processes.
It is likely that both have a part to play, and further analysis of the
case studies could certainly explore this further. But even before
further analysis some significant conclusions for policy follow.
- Research funding should support the full range of disciplines.
There is a tendency in policy circles to try to prioritise between
subject areas. But rather than choosing the disciplines that maximise
impact, the best outcomes will come from supporting research across the
full range, and allowing interactions between diverse areas. At a
national level it might be appropriate to identify priority areas for industrial development, key technologies for the future or important societal challenges,
but a set of research priorities do not follow from that. In fact, the
best strategy for research is one of diversity rather than priority.
- Universities are uniquely placed to deliver impact from research.
Compared with other organisations involved in research and the delivery
of impact from it, universities are the only organisations that have
the integration of different disciplines at their heart. Not only do
they bring together diverse arrays of subject experts, but their culture
is focussed on values – rigour, commitment to evidence, collegiality –
that provide the right environment for the integration of disciplinary
knowledge. Many of our universities have the diversity of excellence
within a single organisation, and more specialist institutions can
easily collaborate with other universities because of those shared
- Knowledge and research integration should be encouraged at all levels.
In order to reap the benefits from disciplinary diversity, the barriers
to working across boundaries need to be minimised. This is especially
the case for interdisciplinary research; it is often suggested that our
research funding and assessment systems don’t support research of this
type. We need to examine this question critically. The case study
analysis shows that there is a lot of activity across the boundaries,
and there is evidence
from the REF that research outputs flagged as interdisciplinary scored
as highly as others. HEFCE, working with RCUK, have commissioned some research
to examine how the UK performs compared to other nations with respect
to interdisciplinary research. We will also examine how REF-submitted
outputs compare to the total UK output.
crosses boundaries, there are also cultural issues. Academic culture is
rooted in disciplinary communities, and it is often, rightly, pointed
out that work across the boundaries needs to come from solid
disciplinary foundations. But we need to develop academic cultures that
recognise and value diversity, and seek to avoid meaningless hierarchies
between disciplines that often get in the way of interdisciplinary
These aren’t new issues for research policy, but the insight from the
impact case studies should put them clearly front and centre in our
thinking. The implications need to inform the deliberations of the review of the Research Councils
that is being carried out by Sir Paul Nurse. The review already has
questions on the balance between disciplines and interdisciplinary
research at its heart, and the case studies provide valuable evidence.
We also need to consider this evidence in the Spending Review that will
surely follow the upcoming General Election. And more broadly, at
whatever level strategies for research are developed, we need to
remember that supporting and enhancing disciplinary diversity is
essential for the delivery of impact.
Dr Steven Hill is Head of Research Policy at the Higher Education Funding Council for England (HEFCE) @stevenhill
Diversity of disciplines drives research impact | Policy Wonkers