Humboldt-Universität zu Berlin - English

Open Science

The term open science is used to describe various practices that aim to increase the accessibility, comprehensibility, and reusability of scientific findings and results in the broadest sense. In the Berlin Science Survey, the degree of dissemination of these practices is surveyed in order to be able to determine trends over time, but also to identify difficulties. 

In the evaluation of goals in science, we were able to determine that open science as a goal in science occupies a middle position in comparison to the other goals to be assessed (see Figure 1), but that the pressure of expectations is perceived to be the lowest here, together with "good teaching" (see Figure 2), and consequently, for the majority of researchers, open science falls to last place in the prioritization of the goals that were compared in the Berlin Science Survey (see Figure 3).

If we ask how important scientists consider the expansion of open science for science as a whole, we find that more than 95% consider it important (see Figure 8).


Figure 8 Importance of expanding open science


So open science does not have an image problem, but the question is whether open science has an implementation problem. To this end, we look below at the results on how widespread open science practices are. 

We asked the scientists how established various open science practices are in their everyday research. It is already known from studies on open science that the most widespread open science practice is open access publishing, i.e., the public and free provision of scientific publications. Here, too, this practice is well established: On average, the scientists state that a good 55% of their publications are available free of charge on the Internet (Figure 9).

Overall, the differences among the disciplines are not extraordinarily large, and even in the humanities, where open access is least widespread, a good 46% of publications are freely accessible. However, the natural sciences lead the way in open access publishing, with over 63% of publications being freely accessible and free of charge. This is not surprising, as the natural sciences already have a long tradition in this area with open science repositories such as arXiv.


Figure 9 Shares of open access publications


Of those who publish open access papers, we asked which of the three most common routes – Gold, Green, or Hybrid Open Access – they use to make their publications available.

Gold Open Access includes all electronic first publications in journals that are free and open access (open access journals). Green Open Access includes all electronic second publications in freely accessible institutional or specialist online archives or repositories, or on websites (referred to as pre- or postprints). Hybrid Open Access, on the other hand, involves making articles freely available to the readership in otherwise paid journals in return for the author and/or sponsor paying a fee.

The average of all scientists shows that the golden path of publishing in a "real" open access journal predominates with 51%. The green path of pre- or secondary publication is taken by 39% of respondents, and the hybrid path is also taken by a good third of respondents (not shown).

However, these average values do conceal differences specific to subject groups (see Figure 10). While the engineering sciences prioritize the Green Route, the Gold Route is dominant in all other subject groups and especially in the life sciences. They, on the other hand, use the Green Route less frequently (only a good 27% of the time). The Hybrid Open Access format is used least frequently by the humanities, at 23.5%. This may also be due to financial reasons, as the DFG funding program for financing open access publications explicitly excludes Hybrid Open Access. Overall, significant dynamics can be expected in the area of open access in the coming years, both on the part of researchers and their institutions due to control policy measures, but also on the part of publishers due to funding issues and developments in contracting between universities and publishers.


Figure 10 Use of different open access formats


Other open science (OS) practices are much less widespread than open access publishing and, moreover, are not relevant in all research contexts (see Figure 11). We assume that open science practices have become routine in respondents' research realities when respondents indicate that they use the practice "often," "very often," or "always."

Making data from one's own research publicly available and free of charge, i.e., data sharing, has become routine in this sense for 26.8% of respondents and is thus the most common of these OS practices.

Code and material sharing, i.e. making study-relevant material from one's own research (code, questionnaires, blueprints, etc.) available publicly and free of charge, has already become part of everyday research for 26.7% of the scientists surveyed.

The inclusion of citizens and/or civil society actors in one's own research (also referred to as "citizen science" in this report) is practiced regularly by only 10.6% of the respondents, whereby the proportion of those who do not consider this practice relevant for their own research is particularly high at 24.7%. The open peer review process has already become routine for 21.2% of respondents.


Figure 11 Frequencies of open science practices


The data on the prevalence of OS practices confirm the suspicion that while OS does not have an image problem, it does have an implementation problem. Despite the high importance given to OS by the scientists, the implementation of OS in the daily research routine of the scientists is given the lowest priority. Low expectations and incentives are probably responsible for this low prioritization in combination with competing goals that are linked to stronger incentives. This could be exacerbated by additional hurdles to implementing open science. We therefore asked to what extent difficulties were seen in implementing each OS practice. Here, all individuals were asked about all practices, regardless of whether they considered them relevant to their own research practice or not.

Figure 12 shows different patterns for the different practices, which are the reason for the rather low prevalence of these practices: While the majority (41.8%) say that they cannot even assess the difficulties in implementation in the case of citizen science, just under 41% of respondents see great or very great difficulties in making data available. This shows that the open science practices themselves are prioritized very differently. More scientists are able to assess the difficulties in implementing open data than in citizen science, which suggests a higher relevance or a higher need for data sharing.

As we have already reported, open access provides a counterpoint to this because it  is now considered to be quite established. Accordingly, 68.8% also see no or only minor difficulties in implementation.


Figure 12 Difficulties in implementing open science practices