The Big Shots of business studies should be role models for Open Science
Stefan Mayer talks about his experience with Open Science
Three key learnings:
- A skill set regarding Open Science will be needed more than ever for working scientifically in future. There are many relevant skills: from the ability to document your own data and analysis scripts clearly and replicably to publishing and communicating them to your own community.
- It pays off für PhD candidates to be courageous and demand an Open Science tutorial or something like it from their supervisor.
- Networking locally for Open Science facilitates access to and handling of Open Science, be it with other PhD candidates, supervisors or the local Open Science office.
You are an active member of the Tübingen Open Science Initiative. How long has this initiative been around?
SM: The Tübingen Open Science Initiative has been around since 2019. At the beginning we asked ourselves how big and where we want to instal this initiative in the first place. Of course it’s ideal if you can instal it at the top level of the university’s structure. At the same time you need people who take care of the small details. So from our exchanges with actors at Tübingen University who interface with Open Science we got the idea to work at the faculty level, i.e. social science and economics. One of the first big events we organised was a Winter School. At first, the initiative was active under the name of “Tübingen Open Science Initiative” mostly in economics and social sciences. But the physicians and psychologists already had a similar intitiative and so we linked up and now have several chapters. Nowadays we try to increase our networking, so we can bring it forward at a higher level.
How did your colleagues react to the start of the chapter in economics and social sciences? Does it make sense to address the topic in a cross-disciplinary way? Or is it more successful to stay within the discipline?
SM: Good question. I think it’s sensible to promote the topic at two levels in parallel. To establish the topic of Open Science as a whole at the university, it makes sense for a central authority to give the signal that Open Science is important for the entire university. That’s the only way to establish Open Science as a topic in teaching and appointment interviews, and it’s the only way to provide funding for things like data management. You should try to place the topic centrally to create an awareness at the management level of universities. At the same time, the real work of course is done in the various sub-initiatives who break it down to the discipline-specific requirements.
How do economic researchers respond to the initiative?
SM: At our faculty in Tübingen we had the concrete case that we added a requirement for candidates to be able to do replicable research into the job description for a new appointment. I thought that was positive. And candidates in the appointment procedure explicitly addressed this. They asked what we understood by this. Which means we need to have an idea of what it means to us. In my perception however, the topic of Open Science doesn’t appeal enough here, nor in Germany or internationally in general. I could wish there were much, much more. Unfortunately there’s not an individual person in my discipline who stands for Open Science – providing the major role model or contact person, so to speak. There are more and more scientists who actively promote this. But right now we don’t have a truly established person whom early career researchers can ask for advice. This proves that it hasn’t become standard yet, alas. Many people still view this critically, which in my opinion isn’t justified.
By “not justified” you mean that many don’t have sufficient information?
SM: Yes, exactly. You first need to know why Open Science is necessary, what is possible and most of all how easily possible much of it is. I often see much ignorance in conversations. A few years ago I gave a talk to some professors from Germany and Europe about “Open Science in marketing”, and the consensus was “Oh, the young researchers can do that. We’ll stick with the status quo.” That is changing by and by. But as in other disciplines, the topic is mostly driven by early career researchers. Some just don’t see it as relevant to their own research, for many different reasons. Sometimes I hear the argument that the data don’t suffice or that colleagues don’t need it for their careers because nobody takes notice of it in appointment procedures. Some people are scared that others might see what they are doing. Of course it makes you vulnerable and that’s why many have a respect for it, because the transparency means that mistakes you may have made become obvious.
Are there entry-level courses for Open Science tools at your faculty to take away colleagues’ fear?
SM: Yes, that was part of the Winter School where we offered workshops for different levels of knowledge. Workshops for complete newbies, but also for people who have looked at the topic before. The target group were PhD candidates, Post Docs, junior professors, but full professors were also welcome. And there we addressed questions such as “What exactly does it mean to pre-register a research?” or “What does it mean if I try to produce a script with R-Markdown that is analysis and output in one?” We had two keynotes (Lisa Matthias, James Heathers) and six workshops, both for qualitative and quantitative research.
Was the offer well received?
SM: We organised this for the first time and were positively surprised by the uptake. The workshops were attended by 10 to 30 people, and for the keynotes we had 70 participants. Considering that we started from zero and spread it very small at the start, we were very happy about that. And there was a lot of positive feedback. The participants also said that they would like to get more of it. For us it was an important experience to see that we’re not staying in our bubble, but that young researchers definitely see this as relevant. You really need to teach the craft in order to take away the fear and the scepticism.
You work in the field of marketing where in the last few years Artificial Intelligence, working with unstructured data or software development skills have been playing an ever larger role. How far does the need for data skills in marketing support your work in TOSI?
SM: That’s indeed an exciting question because in the field of machine learning some aspects of Open Science are widely spread. It’s quite usual to publish the code for papers directly on GitHub (see also https://paperswithcode.com/ ). But also, for example, it’s usual to provide datasets and it’s possible to count them as publications. Reviews from big conferences such as NeurIPS are publicly accessible. So there are lots of connecting points with the topics we address at TOSI. And there are also exciting connecting points in our teaching. In one of our master programmes we have an obligatory module where students must learn everything about a “Data Science Project”: How do I get started? How do I use GitHub? Which machine learning methods are useful in which cases? How can I do a reporting so that other people can work with it interactively and understand what’s behind the data? So this is also about Open Science.
What is the role of Open Science in your field of business studies?
SM: That’s a difficult question. In my perception the role of Open Science is still too small. One of my colleagues has started looking at how many studies in marketing are replicable, for instance. And that number is very small. Pre-registrations are still far from being the standard everywhere. I find that pretty scary considering how easily it can be done. When I write reviews for conference proceedings, for example, I also request that further studies should be pre-registered. Because at conferences you present research that is still “in progress” and hasn’t been submitted yet. I think there are many who still haven’t got their heads around it that you should do this. There are some journals nowadays who at least appreciate it. But it’s still not standard and not routinely required. If I still read in a paper from 2022 “data available on request” I’m often a bit angry. In many cases there is no valid reason to hide or not share data that I have collected myself. I think we have a long way to go yet before this is truly established in business studies. And it would surely help and speed up the development if the “Big Shots” would provide the examples for making Open Science the standard. I believe that more people would go along with it than if only many individual early career researchers do it.
Do you have fellow campaigners or cooperation partners at Tübingen Open Science Initiative to get the topic on the road more broadly in your field?
SM: All who participate in this initiative are very much engaged although they do it on a honorary basis only. But we must link up better with other initiatives and activists, also within the university. There are many interested scientists already, but probably many who have never heard of Tübingen Open Science Initiative yet. So we try to address people with events at different levels. Recently we relaunched ReproducibiliTea Tübingen, informal meetings where we discuss papers concerning Open Science. This journal concept exists now worldwide.
Could you find fellow campaigners in marketing to establish Open Science more strongly in the field?
SM: I believe it would be no problem to find fellow campaigners, because it has just been a topic at the VHB conference. They had a panel about “Data as a production factor in scientific work”, and quite quickly a debate started about what it actually means to have open data. I think the consortium BERD@NFDI alone which is heavily funded will ensure that Open Science gains more ground in business studies. A matter of discussion with fellow campaigners would be whether to try to establish the issue for business studies in general, or if it wouldn’t be better to break it down for the various sub-disciplines. Of course there are areas that concern everyone equally, but then again there are many things handled differently in the various disciplines.
Of course you would need regular forums to discuss this, such as the annual meeting of the VHB.
SM: Yes, the various panels for young researchers last year provided a good starting point. In principle I think it’s almost schizophrenic that on the one hand you teach at university how to do proper research, and on the other hand you don’t comply with these rules in your own research. That’s one of the reasons why I wanted to change things and spent more and more time on it. I believe if you create awareness for Open Science, as I mentioned before, then you will find ever more like-minded people very, very quickly.
The German Research Alliance (DFG) likes to see research output published in Open Access. Members on appointment boards like to see grants raised from the DFG. How much can openness pay off for your own career in the end?
SM: I would recommend to all early career researchers to acquire a skill set regarding Open Science because it will definitely be needed in the future. They should learn how to handle the publishing of data and analysis scripts, but also the skills you need to make your research more accessible. This includes tweeting about a paper or automatically linking a preprint when you’re reporting a paper that isn’t available in Open Access.
Where should early career researchers start who are interested in Open Science?
SM: This has so many dimensions. One of them of course is to acquire the content, and there are many resources for that. There’s an incredible number of Open Science lectures on YouTube that you can watch, blogposts etc. I always recommend Open Science Framework as a starting point. You can see what is around and how it works in principle. In an ideal world there would be an Open Science agenda for the discipline of business studies that you can use to guide you. Another dimension is to find the courage to demand Open Science, to go to your supervisor and to say I want to do this, but I don’t know how. Can you tell me how this works. I can recommend finding this courage to actively demand a training in the field of Open Science. It’s a mix of acquiring knowledge and implementing in practice. At best you infect other PhD candidates, you network and share your experiences. It helped me a lot, because I wasn’t the only one among us who did this. It’s important to have a local contact; they can be fellow PhD candidates, the supervisor or the Open Science office of the university which will hopefully grow in number in the future.
The questions were asked by Dr Doreen Siegfried.
The interview was conducted on April 20, 2022.
About Professor Stefan Mayer
Dr Stefan Mayer is Professor of Marketing Analytics (Tenure Track) at the Faculty of Economics and Social Sciences of Tübingen University. His research interests are product design and visual aesthetics, machine learning and customer judgment. Professor Stefan Mayer is also an active member of the Tübingen Open Science Initiative.