How networking within the Open Science community supports your own research
Dr Caroline Fischer talks about her experience with Open Science
Three key learnings:
- Open Science skills can be a unique selling point in hiring procedures which is increasingly relevant to reviewers.
- Prereview platforms offer an opportunity to invite feedback from peers and to improve the manuscript before the actual review process starts.
- Preregistrations can be helpful in publishing insignificant results.
You have been a Fellow in the programme “Free Knowledge”. How do you become a Fellow and how have you been made aware of Open Science?
CF: I first came into contact with Open Science during my studies through a lecturer who was very much engaged in it. For instance, we worked with open datasets during research-teaching projects, and he also explained where to find open code. As a user of open resources I loved this. I built on this when I taught at university myself and I used a lot of open resources with my students, such as data or OER. In addition, I focused in my research on “Open Government” to find out how transparent the state is. I think it’s paradoxical that we public administration scientists always say we need more Open Government Data and public administration needs more transparency, but we don’t implement this ourselves. These paradoxes in my own scientific work and my positive experience as a student and a teacher are the reason why I wanted to do this myself. And because I have benefited from it, I now want to return something.
What have you learned as a Fellow in the programme “Free Knowledge”?
CF: I was on the lookout for mentoring as a PhD candidate. How do I do a preregistration? What must I consider when I want to publish my data? This is often much more difficult in the social sciences than in the natural sciences. It has been very profitable to get to know the community as a fellow and to join this network. And the network has encouraged me to remain in this area – although there are many opponents and misgivings in my discipline. I was writing my PhD thesis at the time and asked myself if I should make my data openly available, because you make yourself vulnerable in case of mistakes. The community of the fellow programme has fortified me in sticking to my moral compass.
Have you ever regretted your decision?
CF: Sometimes yes, sometimes no. It is a lot of work to make data for studies openly accessible and adaptable. On the other hand it’s my impression that Open Science plays a larger role in application procedures. If you say that you have experience with Open Data, people show interest and try to buy such Open Science skills by hiring.
Are methods and disciplinary culture a subject of discussion in business studies?
CF: In my field, Public Management, a lot is happening methodically. The discipline is opening up towards experiments and taking a more critical position against p-hacking and HARKing. The discussions around replications in psychology have arrived in business studies. There are many people now who reflect methods much more than a few years ago and who question studies. At least that’s my impression and possibly a little unfair to the old hands. I think there’s a lot happening in quantitative research. And in the end it leads to preregistering studies and openly depositing data in order to be more transparent and more reproducible. It’s more difficult in qualitative research, or at least many researchers perceive it thus. But even here help is more and more available such as the guideline for publishing qualitative research data written by Isabell Steinhardt et.al. What’s happening in business studies arises mostly from reflection on methods, in my impression. Many business researchers have realised that it is important for a publication to have a sound methodical backup. In the field of Public Management there are many Q-scientists who believe Open Science is important and who implement it. It hasn’t had much impact yet among chair holders. There are few who actively practise Open Science and I wish there were more openmindedness and courage.
On the other hand, professors discuss topics such as Open Access at the annual meeting of the VHB (German Academic Association for Business Research). In my view business studies are still in the stage of debating Open Access. This topic ought to be over and done with, after all these years. Open Data on the other hand is not really widespread in German business research.
Which positive effects do you see as a Q-scientist in the context of Open Science?
CF: I have profited from the open work of the community and learned a lot. For instance, co-authoring teams arose from the circle of fellows. Writing in these interdisciplinary teams has enriched me. At the meta-level we also made a contribution to the 2020 Heinsberg study.
What I have also noted: when I apply to foreign universities they ask more directly what I have done in the context of Open Science. They also look more closely how you practise science communication and how you interact with society, for instance by publishing in practice-oriented journals.
I think it is also important in the spirit of Open Science to talk about errors and things that do not work in the process of research. This may of course be a sensitive issue for a Q-scientist, but I think this kind of transparency is important. There are more and more people who write about their rejections in the science system. You can share your rejected funding applicatons in forums (e.g. Open Grants). Only recently I took this route myself and wrote about a failed experiment as a methodical failure and what you can learn from this. And the article was accepted in a comparatively new journal, (JOTE Journal of trial and error). I think such outlets are needed in our discipline as well.
How can business researchers best acquire Open Science skills?
CF: In my opinion, every faculty should have a contact person for Open Science, maybe even every discipline. There are enough experts at universities and libraries for the individul aspects of Open Science – for Open Access and data at libraries, for OER frequently at teacher-training courses or in didactical education. Someone ought to build a professional network for this. Ideally, the university management would organise this. It would also be desirable to publish the names of all Open Science experts at the institution so you know whom to ask for which problem. Basically you need a knowledge map for this topic which is not limited to the university, the learned societies would be better suited. If you are a member of the community you know whom you can ask. But for people at the beginning of their PhD thesis it is difficult to find contact persons by themselves, because usually they don’t provide help in an offical capacity. And if there is insufficient time for long researches, the motivation to do open research quickly vanishes.
Will Open Science function only through intrinsic incentives or can we reach the next level in different ways?
CF: Economic research will have taken a giant step if we no longer penalise Open Science and if we honestly and openly name defects in the science system and start discussing them. We also urgently need a different culture of error. We must take more chances and risk errors. Making errors, finding errors and learning from them is a good thing.
Of course it makes sense to provide impulses so that researchers become curious about Open Science. But after that they need to do it under their own steam. What is needed is the acknowledgement of Open Science in career paths. If you only work with rewards, intrinsic motivation is lost. If I only do Open Science to earn a bonus, I’m no longer an open scientist. In my view this leads to insufficient reflection and I worry that more data would be published that really shouldn’t be – either because the test subjects didn’t give their consent or because the data are not good (e.g. no validated scales). What is really needed is the acceptance of Open Science in career decisions, e.g. the inclusion of altmetrics, data publications or the production of OER, a least as B or C criteria in application procedures.
Would you say that the science of the future is more teamwork?
CF: Exactly. Honestly, that’s what I think anyway. You need a lot more diversity in the research team starting with methodical expertise. But you also need so many additional skills when it comes to science communication. If we talk about incentives for Open Science, I would want universities to offer more experts who take care of science communication or data processing for me. It’s just so much effort to prepare the metadata or a good code book or to rewrite your findings in ways that ordinary people can understand. I don’t think that the public relations offices in universities do this sufficiently, at least not in a particularly proactive sense. It’s my impression that this kind of labour division is more widespread abroad, especially in departmental structures.
You stated in the Heinsberg paper that prereview platforms must be developed further for Open Science. What do you mean by this?
CF: About the Heinsberg study: the scientists made their findings public in a press conference before the review process. Of course, in such a crisis situation it was important for society that results are communicated. But if you have no second or third opinions on the design and the results of the research, this is fraught with difficulty. There’s no quality control. The aim here should be to enter the review process earlier. It’s possible to publish preprints, e.g. on the Open Science Framework, and make them available to the scientific community before the formal review process starts. Then the scientific community can try to oganise the review process itself instead of waiting for the publishers to take care of it, especially when speed is essential. Platforms such as https://outbreaksci.prereview.org or https://prereview.org or in my field Open Science Framework or SocArXiv offer an opportunity to accelerate this feedback process. In principle it’s comparable to a conference where I can get feedback from my peers. But a conference is held once a year and a prereview platform can be used all the year round. If speed is vital and if I want to present my findings to the public before the review, such platforms are a great option. And of course they offer much more transparency than a closed conference.
How do you implement Open Access?
CF: I always try to publish an open preprint in parallel if the publisher allows it. Usually I use my university’s repository or the OSF. If preprints are not allowed, I also pay Open Access fees. But so far I have refused to go beyond the mark of 1,000 Euros that some publishers demand. In such cases I prefer postprints after the embargo on the article has ended.
What is your experience with preregistrations?
CF: My experience has been very good. Only recently with a paper where the result was null, meaning there was no effect. And I have always been told that publishing null results would be hard. But it was no problem at all. I submitted it and received super encouraging reviews. The reviewers referred to the preregistration und considered the null results as valuable. This has taught me that it is useful to have a preregistration up your sleeve if things go wrong or there are no results which can happen quite often with experiments in the social sciences. You can show you didn’t manipulate your data until some result showed. This has helped me a lot. I generally find that preregistration is useful to control if I have sufficiently thought through my research design. Preregistration forces me to discipline myself towards an ideal research process, to plan precisely before collecting data: what are my hypotheses? How exactly will I collect data? How will I analyse my data?
Where do you do this, at Open Science Framework?
CF: Exactly. They offer various templates I can work with. I like the AsPredicted one a lot, it’s not too detailed, but offers checks for everything that could be relevant. I’ve got some preregistrations lying there right now. One of the advantages is that I can put an embargo on them so they only become visible when I submit a manuscript; or I can deposit preregistrations blindly. I can then declare them in the review process and it still allows a double-blind review. It’s a lot of effort, but I am huge fan.
Where do you publish your research data?
CF: So far I have published research data at Open Science Framework, too. Here people from my discipline can access them, not only scholars from my own university. And it has few obstacles. At OSF I can do everything self-directed and without lots of coordination. I don’t have to clarify with a data centre what rules should be followed during data collection. I have to admit, though, that I didn’t need special access controls for my data because so far I haven’t worked with sensitive data yet where that would have been necessary. I always link the data to the corresponding articles so that it’s clear what belongs together. Transparency and reproducibility are important to me.
Do you rework your research findings for a non-scientific audience? If so, in which form and what’s your experience?
CF: I try to publish topics that could be interesting to practitioners in the relevant trade journals. I have quite a few publications in such magazines by now. I also give talks or workshops at training events for practitioners in public administration where I include my research findings. This creates a different kind of exchange and I get a different view of problems that must be solved in the real world. I am also active on the “Ring a Scientist” platform. There teachers can look for experts on certain topics and invite them into their schools. My discipline is probably not the most relevant topic in classes but I have had some invitations already. Mostly I was asked to explain how science can be a profession and how social science works. This is fun and it’s a different audience; their questions are fresh and they also question well-worn routines or things that you have long since resigned yourself to.
What are your tips for economists who want to enter the world of Open Science?
CF: I think the most important thing is to find a community and to join it. The Fellow programme sadly has been ended (for now), but Stifterverband for instance offers a similar programme. Just take a look and get in touch with the big players, e.g. Volkswagen Foundation, Stifterverband, Wikimedia etc. I would try to find institutional players who support projects. At the same time try to find out who is already doing this in your own field of research, who is using open research methods. These people are usually highly motivated to pass on their knowledge. You could look who always publishes data with their articles and simply contact them. The Open Data or Open Code badges show this clearly. In my experience the community is very open to start an exchange through sharing tips and experience.
Many universities have formal mentoring programmes where you can choose your mentor. You could pick someone from the Open Science community for this.
All in all, there is not the one decisive group you can turn to. You have to invest into some research effort to find someone to contact. But I would always take this step before trying to do it all on my own.
The interview was conducted on June 3, 2021.
About Dr Caroline Fischer
Since October 2021 Caroline Fischer has been assistant professor at Twente University in the Department of Public Administration. Her research is focused on knowledge management and organisational learning, human resources management in the public sector, the digital transformation of public administration and its effects on staff and citizens.
Caroline Fischer received her PhD summa cum laude in 2020. She had a scholarship for the Junior Teaching Professionals Programme of Potsdam Graduate School. In 2017/18 she was a Fellow in the Free Knowledge Programme of Wikimedia Deutschland, Stifterverband and Volkswagen Foundation. She writes about Open Science in public administration research in her blog.