The Leopoldina as a facilitator of change towards Open Science
Interview with Leopoldina President Prof Dr Bettina Rockenbach
Photo: Anna Kolata for the Leopoldina
Was there a particular moment when you first became aware of the topic of open science? When were you first sensitised to issues of transparency and openness in science?
BR: To be honest, I can’t name a specific point in time. Until a year and a half ago, I was Vice Rector for Research at the University of Cologne for over eight years. In this role, open science was a key topic – regardless of my own research.
Did you become aware of the topic of Open Science through colleagues, or did you realise yourself at some point that more openness in research would be beneficial?
BR: Open science offers clear advantages, especially in experimental research. This is well known in economic research. I work empirically myself and collect data – so it makes sense to work with the data of others, to check my own analyses against it and to contrast it with my own results. At the same time, I believe it is important to make my own data accessible to the scientific community – for reasons of transparency, efficiency and traceability. It’s about disclosing the analysis steps so that results can actually be replicated. This topic was taken up early on in psychology, somewhat later in economics. But here too, experimental research has increasingly recognised the relevance of replication and data access.
You are still relatively new to the office of President of the Leopoldina. What importance do you attach to the topic of Open Science for the work of the Academy – particularly with regard to its role within the scientific landscape?
BR: The Leopoldina has repeatedly and clearly positioned itself on the topic of open science, both within the framework of the Alliance of Science Organisations and in its own statements. From the Academy’s point of view, this is a cross-cutting issue of great relevance for the future organisation of scientific practice. Open science promotes transparency, efficiency and accessibility. The reusability of data accelerates scientific progress and reduces redundancies. At the same time, open access creates new opportunities for participation – especially for researchers at less well-equipped institutions worldwide. Open science is therefore a central component of an open, responsible scientific culture.
As Vice-Rector at the University of Cologne, you were actively involved in the development of Open Science principles. To what extent does this experience feed into your current work at the Leopoldina?
BR: At the University of Cologne, we have developed an Open Science guideline aimed at anchoring the corresponding principles institutionally. The work at the Leopoldina differs in that the Academy itself conducts virtually no primary research. Its activities focus on science-based statements based on the expertise and publications of its members. These statements reflect the current state of research, take up developments such as open science and are of course published transparently. The structural proximity to scientific discourse is therefore maintained, even if the role is different to that of a research-based university.
You have just mentioned the Leopoldina’s statements. Recently – about a fortnight ago – a discussion paper on the direct funding and evaluation of scientific journals was published. What role does the Leopoldina see for the transformation of existing publication models towards open access?
BR: The discussion paper addresses a central problem that has existed for some time in scientific publishing. Scientists make a considerable contribution to the quality assurance of scientific journals by writing articles and peer reviewing, and they do so free of charge. At the same time, they are confronted with high costs: either via subscription fees or via author fees in the open access model – the APCs. Against this background, the question arises as to whether this system is still up to date – especially as editorial and technical processes can be organised much more efficiently today than in the past. The idea developed by the working group is therefore aimed at an alternative financing model. The Leopoldina expressly supports this discussion and sees itself as a driving force for the further development of the scientific publication system in terms of openness, fairness and sustainability.
In June, an initial symposium was held to further develop the model outlined in the discussion paper. If this results in concrete proposals, does the Leopoldina see itself in the role of actively representing these to political decision-makers such as the BMBF? Does it see itself as a driving force in this transformation process?
BR: The Leopoldina certainly sees itself as a driving force and supporter of this discourse. Our task is to introduce science-led ideas into the public and political debate – this explicitly includes the transformation of publication models. However, the actual implementation, particularly in financial and administrative terms, is not the responsibility of the Leopoldina. We have neither the resources nor the institutional structures to support or manage such a model ourselves. Other players would be responsible for this – for example the German Research Foundation or the Federal Ministry of Education and Research. Our role is therefore primarily to initiate such initiatives, underpin them scientifically and feed them into suitable political contexts. Operational implementation must then be carried out by the institutions designated for this purpose.
Does the Leopoldina see itself as a driving force in this context – as an institution that formulates a vision, clearly identifies the need for action and brings together specific players from the scientific and publishing system?
BR: The Leopoldina sees itself as an organisation that formulates science-based recommendations for action. Whether and how these are implemented depends on the respective addressee: In the case of policy advice, it is the politicians who decide; in the case of society-related topics, it is the public or individual institutions. In this case, it is an internal scientific issue. The scientific community itself must consider the proposed model to be viable and sensible – and be prepared to implement it if it is more convincing than existing structures.
If we take the Leopoldina paper on the publication system as our starting point, we are at the end of the research process. It starts with an idea, followed by research – be it experimental, empirical or theoretical. Years often pass before results are published. Are other groups working on concepts aimed at upstream phases of research?
BR: I think that the topic of publication systems – particularly in the context of technical developments and the switch to open science – is already being strongly driven by the specialist communities themselves. New approaches are not developed in isolation, but through dialogue: at conferences, in expert discussions, in international discourse. Scientists contribute their ideas at an early stage, discuss them and develop them further. In this respect, the idea that research results remain internal for years is increasingly outdated. The shift towards greater openness does not just begin with publication – it permeates the entire process.
Where do you currently see the biggest challenges in science communication? Scientific texts are often highly complex – even specialist colleagues from neighbouring disciplines need context. What opportunities and difficulties do you see in communicating scientific content to a wider audience?
BR: Science communication is a key issue for the Leopoldina. It is about preparing our statements in such a way that they can be understood and categorised in society. This is fundamentally important – and in view of increasing scepticism and hostility towards science, it is becoming even more important. We need services that make it clear what science does and what contribution it can make to guiding social or personal decision-making processes. This applies to both individual assessments and political judgements. This is precisely where we are trying to start.
You are a behavioural economist with many years of research experience. What perspective does your discipline bring to the open science transformation? Where do you see behavioural levers that have so far been little used?
BR: In principle, I welcome the sharing of research data – especially in a way that enables replication. One aspect we haven’t talked about yet is preregistration. It starts very early on in the research process. From a behavioural economics perspective, I see opportunities here, but also risks. Personally, I am concerned that overly strict preregistration practices will result in creative and unexpected findings losing visibility. If only what was precisely planned beforehand is published, those results that were not anticipated but are highly relevant scientifically could be lost. Gaining knowledge in research also thrives on surprises – these must not be suppressed by formal requirements.
As I understand it, proponents of preregistration do not demand a rigid commitment to the original plan. Corrections or additions are possible at any time, including subsequent registration. The aim is to avoid so-called p-hacking – i.e. the targeted “fishing around” in the data in which non-significant results are suppressed and only suitable results are published. Pre-registration should create more transparency: The research question, the data basis and the methodological approach are disclosed in advance in order to categorise the search for significance.
BR: That’s true – but in practice you still encounter restrictions. Sometimes reviewers complain: “This analysis was not preregistered.” And then you have to explain: No, because we didn’t expect this finding – but that’s exactly what makes it interesting. The scientific discovery process is not always linear. Surprises are often the driving force behind new insights. That’s why I advocate a sense of proportion: preregistration is an important instrument for transparency, but it must not lead to the unexpected no longer having a place.
Looking back on your time as an active economist, were there any examples from your own research practice where the open exchange of code, protocols, data or publications was particularly enriching for you?
BR: Yes, we had a highly cited paper in Science with, as we found in our own replication studies, very robust results. We built on this ourselves and carried out further studies. A few years later – I think it was about ten to fifteen years in between – we were approached by an international consortium. The colleagues asked for our data, the experimental protocol and the code to replicate the study in seven countries. The very good news was that our results were reproduced, even after many years and by different international groups. That was really impressive and also very pleasing.
But I would like to emphasise one thing: In the behavioural sciences, you have to be careful with replications. We study human behaviour – and that changes over time. If a result can no longer be replicated after 20 years, this does not necessarily mean that it was wrong at the time. It is simply not a physical law that applies regardless of time and context. In our case, the results were stable – but this is not always the case. The discussion about a replication crisis must therefore be differentiated: There are questionable findings with samples that are too small, but there are also findings that change because contexts change.
I would like to look to the future. You must have supervised doctoral students yourself. If you were to give recommendations to young economists or scientists without previous open science experience today – what would you advise? What should they pay attention to when dealing with topics such as openness and transparency in research?
BR: There are now clearly established standards in our discipline, ranging from pre-registration to the provision of research results. Whether you can afford the additional fees for an open access publication depends on the respective framework conditions at your own university, especially if the journal itself does not offer open access. But the other steps have largely become mandatory. Those who do not adhere to these standards have little chance of publishing in renowned journals. In this respect, the framework for action is clearly defined by the leading journals.
In 2023, we conducted a study on open science in economic research and analysed various practices – from registration to the sharing of research data. My impression is that there is still a certain amount of uncertainty in the context of open science. There is often concern that someone might discover a coding error or point out another mistake, for example. The fear of criticism seems to be greater than the willingness to take advantage of the opportunities offered by open scientific dialogue.
BR: That can certainly play a role. Another point is certainly: Have I already analysed the data as I had planned? Or are there still questions that I want to answer myself before others do? In such cases, researchers may be reluctant to share their data at an early stage. However, leading journals are increasingly imposing requirements: for pre-registration, for disclosure of the analysis processes and for specifying a repository for the subsequent use of the data. Even if this has not yet been implemented everywhere, it will become standard in the future. At the same time, current developments – for example in the USA – show that one should not rely solely on the fact that data is securely stored in a repository. Data can be lost or researchers can lose access to their own datasets. It is therefore advisable to keep additional redundant backups.
Thank you very much!
*The interview was conducted by Dr Doreen Siegfried on 20 May 2025.
This text was translated on 10 June 2025 using DeeplPro.
About Professor Dr Bettina Rockenbach:
Prof. Dr Bettina Rockenbach has been President of the German National Academy of Sciences Leopoldina since 1 March 2025. She is the first woman and also the first economist to head the Academy. Rockenbach is Professor of Experimental Economics and Behavioural Research at the University of Cologne and Senior Research Fellow at the Max Planck Institute for Research on Collective Goods in Bonn.
Rockenbach studied mathematics, economics and computer science at the University of Bonn and completed her doctorate in 1993 under Nobel Prize winner Reinhard Selten. She also completed her habilitation in Bonn in 1999. Before her appointment to Cologne in 2011, she was a professor at the University of Erfurt. She has been a member of Leopoldina since 2013 and was Vice Rector for Research at the University of Cologne from 2015 to 2023.
LinkedIn: https://www.linkedin.com/in/bettina-rockenbach-8a888a366/