Science is too important to ignore Open Science
Dr Ulf Steinberg talks about his experience with Open Science
Three key learnings:
- Early career researchers show a keen interest when experienced scientists share their experience with, and approach to, Open Science in practice-oriented workshops.
- As with every big change project in business, it is also indispensable in science to approach the Open Science transformation at various levels and with different measures.
- Graduates who know how to handle data are well-positioned for future challenges.
You have been active at a structural level in the commission for Open Science of the German Psychological Society. What was your work?
US: I’m still active in the commission on a voluntary basis. The commision for Open Science was established by the German Psychological Society. The task was to accompany and evaluate the introduction of recommendations for research data management, but also to edit them and in a wider sense prepare, follow and help shape developments around Open Science for the Psychological Society. But the commission also functions as a point of contact to other learned societies and institutions. The commission for Open Science grew from working on the recommendations for research data management. Originally it was a working group around Felix Schönbrodt, Andrea Abele-Brehm and Mario Gollwitzer. I joined after the first version of the recommendation had been published and an evaluating study was planned. Then things gathered speed and by now we have published a second version which addresses farther-reaching aspects of research data management. Now we are planning more activities regarding exchange. We have been having more talks with the ethics commissions in psychology and we have a series of Round Tables in planning where we want to have more exchange with other actors in the field. It’s important to us to promote networking here.
How can you create structures in a learned society that actually lead to changes in behaviour?
US : In the commission we encounter researchers who are just now entering the field and do their first preregistration and probably don’t know yet that there are obstacles to overcome; who are now learning that issues of data protection, copyright and research ethics must be considered. But we also meet with much and keen interest. It has moved us, the commission, last year to offer a workshop, very much practice-oriented, where experienced colleagues shared their experiences and approaches to Open Science. We have entered a very productive exchange here. The workshop met with much interest especially among younger colleagues. As with every big change project in business, it is also indispensable in science to address the topic at various levels and with different measures. In my opinion, you have to learn about Open Science because in future funders will require you to work according to these principles or else there will be no money forthcoming. My other perception is that a few years from now, nobody will have the time to read research papers that are not open. In economics, too, people will ask themselves if they’re willing to read papers that are proprietary and thus potentially less verifiable, or if you prefer to spend your limited time on reading papers where you could ascertain yourself whether the findings are accurate or not.
What is the best starting point for establishing Open Science in a whole discipline? What’s the best way in your opinion?
US : There are strong efforts in psychology to address both support and infrastructure. ZPID – Leibniz Institute for Psychology Information (ZPID) is very active in providing infrastructures for psychology research. They have created a very substantial range of offers. A useful supplementary measure for infrastructures are workshops, especially for those scientists who are in mid-career. It’s important here to send the message: “You have ben trained in a certain way of doing scholarly work and you are involved in developing the discipline and its methods. But now there’s a need to rethink scholarly work. Please follow this path, and do it sooner rather than later.” In the long run, I think change will happen because young scientists are being introduced to this topic very early. In psychology, for instance, there are practical tutorials in empiricism where students learn how to do research. Many universities have firmly embedded this in their curricula; the first universities have incorporated this into the PhD degree regulations in psychology; it plays a large role in tenure track agreements, and it has also entered the appointment procedures; at LMU in Munich it’s now a standard paragraph and part and parcel of appointment procedures. All these are measures that must become widely accepted over time from the bottom up. But I think at the end these measures will have the intended effect. A few trailblazers serving as role models will have large imitation effects in this. In the commission we try to identify such trailblazers and to bring them into exchange with other actors. There are many such actors after all, such as NOSI or German Reproducibility Network and in other countries, too.
Where do you see connecting points for economists or their learned societies for collaboration, for exchange, and for taking away something they can use for themselves or the institution?
US: I believe there are many different levels on which you can enter into productive discourse with others. That’s characteristic for the work of the commission for Open Science. When we discuss problems, it concerns some scientific areas more than others. For example: if we say it would be desirable for researchers to share their research data as openly as possible, then we would run across a number of problems and questions that we haven’t been able to solve completely yet, but which would concern economic research just as well as psychological research. For one thing there’s the question of copyright: if you use other people’s data in order to have a sufficient data basis in the first place, you may possibly not be allowed to share them at all, or only part of them, or only under certain conditions. So what can you do? What are the options? In our revision of the recommendations for research data management we have made a proposition regarding this problem, namely the publication of data within different types of access; types that can be represented by infrastructure providers and framed in licence and contract laws. It was my understanding that the replication problem is rarely talked of in economics. Instead there is a reproduction crisis or at least questions about the reproducibility of study findings. It’s more a question of whether other researchers produce the same results when working with the same dataset. It doesn’t have to be the case, that happens both in economics and psychology. Although there hasn’t been heard much about crises in economics, it’s true that false research findings have implications for economic policy decisions and thus for human life. Bearing this in mind, it would be useful if the data on which the article is based would also be supplied, so you can verify the computations. This would be an important action under many aspects. Not only because of the policy implications, but also for methodology discussions. Society, which funds many research groups, should get a clear and robust picture of a research subject in return for its outlays.
Where can economists go if they want to copy something specific from psychology? What can you recommend?
US: One point of reference is the website of the German Psychological Society, currently under reconstruction, where there is a collection of links. There’s the Leibniz Institute for Psychology Information which also has a comprehensive collection of resources and in the English-speaking world there’s Open Science Framework, which is cross-disciplinary in its approach.
We have noticed that in economics it is often unclear what the term Open Science encompasses. What are the concepts you are working with?
US: We have used Open Science for the entire dicourse because concepts such as transparency or robustness have their own connotations. Even if suddenly we all shared data and analysis scripts etc., this would not necessarily comply with the aspect of a transparent science communication because a certain contextualisation would be lacking. Of course you could say there are scientists who can interpret this, but it wouldn’t really be truly transparent communication. The term “robustness” is mentioned again and again, especially in connection with the replication crisis. But we have a clear focus on the topic of Open Science which rests on several pillars, such as Open Data, Open Materials, Open Access. verschiedenen Säulen zusammensetzt. Dazu gehören Aspekte wie Open Data, Open Materials, Open Access.
You are now a scientist in a non-academic field and work for a management consultancy: how do you get access to literature or data? And what would be your ideal concept of organising exchange with other researchers at a material and communicative level?
US: When I was still at university I opened accounts with several state libraries because all university libraries in Germany have their limits. As long as all-inclusive agreements with publishers are not concluded nationally, even scientists will have to depend on creating accounts with state and other libraries. That is one option of accessing literature hidden behind paywalls. Although the share of articles published in Open Access grows perceptibly. In many cases you can download these articles directly from the databases, so that one of the barriers between institutional science and science conducted outside of business has fallen. The same would apply to data. An issue that crops up repeatedly in connection with research data management is data cemeteries. Are truly all data of interest for posterity?
Isn’t that rather a luxury problem?
US: Yes, but it is an argument of people who see the change towards Open Science critically. Reuse is an important issue. In our company we have data that would be highly interesting for leadership and management research. Here you need to consider how business can be incentivised to open up and participate in the scientific discourse. Business on the other hand is interested in using analyses and findings for themselves, for instance to have good change management.
Which future skills regarding scholarly work do students need who want to work for management consultancies?
US: My experience is based on the fact that I studied at a very large faculty of business studies; accordingly I maintained exchange with students and supervised final theses. I think economics could put more focus on methodological teaching when it comes to asking: How are data created? What can these data reveal? How can we interpret these data and maybe reuse them automatically? Compared to psychology or the social sciences, it is more important in business studies to find your individual direction as a student. Some universities have established initial progammes that are focused on topics such as People Analytics or AI for Business. Basically I would tell all young people: If you know how to handle data, you are well-positioned for the future.
How can young scientists engage for Open Science?
US: There are many opportunities. The most important thing is that you want to engage, and to take the first step. The most direct path is to contact scientists already engaged in Open Science because this is a wide-spread movement. In my view, all colleagues are very open and engaged. So just write an email and ask what you can do. Many of the Open Science initiatives at universities have sprung up in this way. And I know that the network of Open Science initiatives tries to establish a system of sponsorships. So if you have absolutely no idea of where to start, there are structures and points of contact for sharing experiences.
Questions were asked by Dr Doreen Siegfried.
The interview was conducted on November 18, 2021.
About Dr Ulf Steinberg
Dr Ulf Steinberg is a manager at the consultancy Manres AG in Berlin. Until 2021, he was research assistant at the Chair of Research and Science Management of TU Munich. He is a psychologist who gained a PhD in business studies in 2019 with a thesis on the influence of power on behaviour in organisations and the digital transformation of leadership and management at TUM Management School. Dr Ulf Steinberg is also a systemic coach working with upper management levels. Dr Ulf Steinberg is a member of several professional associations and a member of the commission for Open Science of the German Psychological Society.