Open science is a spectrum and we must push for greater inclusivity

Open science has made substantial progress over recent decades, building on the progress of the open access movement.

When the Center for Open Science (COS) was established in 2013, much of the discussion was about what open science was, why we needed to practise it and how it could be advanced. The conversation has shifted dramatically since then, with the inclusion of open science in policies and recommendations from the European Union (Horizon Europe), the White House Office of Science and Technology Policy (Federal Year of Open Science) and UNESCO (Recommendation on Open Science).

Yet challenges persist. Open science has the potential to make the scientific process more transparent and inclusive, which will accelerate discovery and innovation and drive more equitable outcomes. This can occur through incorporating under-represented or under-recognised perspectives in the research process, as well as enabling participation in and access to research.

Open science will also expose questionable research practices. For example, the sharing of data, code, analysis plans and results in papers published by Francesca Gino led Data Colada, a trio of scientists, to suspect fraud after further investigation. When cases like this occur, it can lead some to perceive science as less trustworthy. However, the more open the process of science is, the more others can engage, which includes maintaining healthy scepticism and scrutiny of research to enable good science instead of just more science. This is a feature of open science, not a bug – and, ultimately, it will boost trust in science, rather than undermining it.

That isn’t to say that all scientists embrace openness or scrutiny. Gino herself launched a libel case against Data Colada and Harvard University, her employer, which was dismissed in September. Moreover, with the best will in the world, we cannot simultaneously open all parts of the scientific process, particularly when ethical or legal considerations need to be balanced (around privacy, for instance). This means open science is on a spectrum. There is more that needs to be done to understand, experiment and advance open science through this lens.

For example, the company Meta is partnering with COS on a pilot programme to share certain privacy-preserving social media data with researchers to study topics related to the well-being of teens and young adults. This pilot project is leveraging aspects of open science, such as Registered Reports – a publishing format in which the research methods and proposed analyses are peer-reviewed before the data is collected – to promote rigour and transparency. And while the social media data will not be publicly available, the rest of the scientific process will be made as open as possible.

The aim of this pilot is to enable collaboration between academia and industry to identify practices for sharing proprietary data among the research community that navigate through the complex and multifaceted risks (commercial, legal, ethical and reputational) that currently disincentivise such sharing.

If successful, the lessons learned from this pilot can provide a mechanism for enabling more credible research by making the process and outputs as open as possible and as closed as necessary.

The opportunity to make open science approachable to as many individuals, applications and use cases as possible also means being mindful of who is engaging with open science and who is not. We have been maintaining the Open Science Framework (OSF) since our foundation and have seen continued growth in the number of researchers engaging in preregistration and sharing data, materials, protocols and code.

Our users are from all over the world, bringing a diverse geographic representation from the US, the UK, Europe, Canada, Australia, Brazil, China, Indonesia and India. What is more impressive, however, is the number of individuals who visit content on OSF, with more than five million unique visitors so far in 2024 and 18 million views. This demonstrates the value of open science, making the research process, and the outputs and outcomes of that process, available to a wide audience.

At the same time, we know there remains much work to do to make open science more inclusive. One way to do this is to broaden the conversation and terminology, such as discussing “open scholarship” as opposed to “open science”. This subtle but significant change enables a wider community of researchers, such as those in the arts and humanities, to participate, and it defines how practices such as data sharing and preprints can be started, scaled and sustained.

Similarly, advancing and rewarding open practices needs to include perspectives from countries around the globe. Enabling and rewarding research progress is already a challenged endeavour, with the overemphasis on research publications as an indicator of success favouring better-resourced researchers and locations. As more of the research process is exposed and acknowledged, the readiness and resources of researchers and their institutions need to be considered. If open science becomes a race to be the best prepared, this risks merely exacerbating global inequities.

The progress made since 2013 shows that open science is no longer niche; it’s becoming the norm. But for it to reach its full potential, we must push for more inclusivity and transparency across all regions, disciplines and organisations in the research ecosystem. By fostering a global research culture that values transparency, we can build trust and ensure research benefits everyone.

Tim Errington is senior director of research at the Center for Open Science, whose many different funder partners include the John Templeton Foundation, the Templeton Religion Trust and the Templeton World Charity Foundation.