With UK universities under severe financial pressure, you might question whether mandates to make research as openly available as possible are still justifiable. However well-intentioned, would new rules to make data and outputs more accessible represent an unacceptable burden on a research system struggling to stay afloat?

Viewed from a narrow, short-term perspective, the answer might be yes. But leadership is about maintaining a long-term strategic vision while overcoming short-term challenges. Those of us advocating for open research practices need to articulate their wider importance – particularly as a Research England pilot study considers whether universities might be rewarded for their open science commitments as part of the People, Culture and Environment section of the next Research Excellence Framework (REF).

There is both a moral and selfish case for open science. The moral case is simply that, as far as possible, publicly funded research should be accessible to those who ultimately fund it. Not just the final research paper but, ideally, the intermediate research outputs we create along the way, such as study plans, data and analysis code.

The selfish case has several aspects. One is that transparency can drive research quality, partly by offering greater scope for scrutiny and error detection. We experienced this ourselves recently – a University of Glasgow MSc student downloaded one of our datasets (as a requirement of their course – there is pedagogic value in open research, too) and spotted a mistake, which we were then able to correct.

黑料吃瓜网

Embarrassing for us? Yes – no one likes errors to be found in their own work (fortunately it was a minor one, a typo in the data dictionary). But our anxiety to avoid further embarrassment incentivises us to develop internal processes that catch as many of our own errors as possible before we complete a project.

Transparency can also improve efficiency. To take one example, sharing data and code in such a way that anyone can understand and use it means that we ourselves?will be able to reuse it when we return to it years later. As the quote goes, our most important collaborator is ourselves from six months ago, and that person doesn’t answer emails.

黑料吃瓜网

As a Wellcome Trust colleague put it, one of the great things about code being open is that it prevents duplication of effort: “One person can code for a specific situation and then somebody else can…with their permission, use that code rather than having to write it from scratch.” But this only works if the data and code is well curated, with proper annotation, a data dictionary and good conventions on naming variables and files.

Productivity gains will be maximised if such good practices are embedded as early as possible in people’s careers – even if this requires an upfront investment of time and other resources. And these working practices may make institutions more attractive places for investment – particularly from industry – and their researchers more attractive to potential employers.

Private-sector R&D organisations – unlike universities – typically have well developed quality assurance frameworks to ensure the integrity of the data they generate. This is because they are generally more incentivised to get the right answer to their question – the best way to make money as a pharmaceutical company, for instance, is to make a new drug that actually works.

Of course, private industry works in a more closed environment than academia does, with firms anxious to keep competitors from cashing in on their own R&D efforts and stealing their IP. But much academic research can’t be made fully open either, for ethical reasons. Either way, the same principle applies: the more that each step in the research process is documented and scrutinised, internally or externally, the less scope there is for error.

黑料吃瓜网

Open research practices can therefore underpin the development of quality assurance frameworks that make working practice in academic institutions more compatible with industry without losing what is distinctive about academia.

Open science can also promote creative thinking about who can benefit from and use the resources created – and beneficiaries are not limited to other researchers. Anyone with an interest in how science is done can learn a great deal from it. For instance, we are working with secondary schools in Bristol to help teachers and pupils use our data deposits and other related study materials in psychology A-level classes, bringing the subject to life.

Creating such direct links between our research activity and local schools can contribute to fulfilling the civic mission on which UK universities are increasingly focused. Moreover, demystifying what universities do could help us achieve our widening participation goals.

Finally, open science can promote greater cross-sectoral collaboration, not only in research but in training. Whatever the incentives the REF may provide to steal a march on competitors by ploughing your own furrow, open science can be more effectively embedded collectively. Developing common training in open research practices, for example, is not only cost-effective (via economies of scale) but also simply better; skills developed will be interoperable across institutions, reducing friction associated with researcher mobility.

黑料吃瓜网

Indeed, open research is just one example of how the sector can and should look beyond the unfortunate zero-sum competition?that exercises like the REF can create. As finances get ever tighter, we can only achieve a thriving and sustainable sector if we think in this more holistic way.

is professor of biological psychology at the?University of Bristol and co-founder of the UK Reproducibility Network. Neil Jacobs is head of the UKRN Open Research Programme, funded by Research England, which aims to extend and embed open research practices across the UK sector.

黑料吃瓜网