Cooperation from the perspective of the scientists
So far, we have looked at the extent and type of research collaborations already established in the Berlin research area. In order to be able to explore university policy measures for strengthening collaborations in the Berlin research area, it is essential to integrate the experiences and assessments of the scientists themselves into the evaluation. We approached this from two angles: On the one hand, we wanted to find out how important collaborations are in scientific work and how pronounced the pressure is to cooperate, especially in an interdisciplinary way. On the other hand, we wanted to measure the quality of the cooperative relationships on the basis of the scientists' reports on their experiences, i.e., what had worked well and what had worked less well in previous collaborations. The general attitudes towards cooperation and the subsequent experiences provide a good picture of how the scientists themselves view the topic of "cooperation".
The Importance of interdisciplinarity in daily research routine
In recent years, the topic of interdisciplinarity has been very much in focus. Although by no means all collaborations are interdisciplinary (cf. Figure 12), nor should they be, such science policy imperatives can certainly exert some pressure on scientists. While the Berlin University Alliance aims to promote collaborations in general, the science policy imperative to carry out projects in as interdisciplinary a manner as possible has existed for some time. Furthermore, research funders have been specifically promoting interdisciplinary projects and alliances for some time.
Respondents to the Berlin Science Survey were asked to assess the importance of interdisciplinary collaborations, the perceived pressure of expectations, and the prioritization of their own research practices. Figure 22 shows that a good two-thirds of respondents reported that "interdisciplinary cooperations" "should be an overriding goal" or even "should be the highest goal" within science. At the same time, more than one-third of the respondents felt "high" or even "very high" pressure to cooperate in an interdisciplinary manner. Finally, in their own scientific practice, this goal was given a "no" or "low priority" by 51.6% of the respondents and a "high" or even "highest priority" by 58.4%. The prioritization of various goals in one's own work was driven both by the setting of one’s own goals and values and by the pressure of expectations.
Figure 22 Interdisciplinarity: goal, pressure of expectations, prioritization
Figure 23 presents the responses in relation to the other scientific goals surveyed. It turns out that interdisciplinary collaborations – although considered quite important – were nevertheless seen as less important than most of the other goals. Only the “publication output” and the “social usability of research results (societal impact)” were considered less important. At the same time, the pressure to cooperate in an interdisciplinary way was rather low, although higher than for “good teaching” and “open science”. At the same time, interdisciplinarity in one's own practice was given a lower priority, as was “good teaching”. Only “societal impact” and “open science” were prioritized even lower in everyday practice.
Figure 23 Importance, expectations, and own prioritization of goals in science
Quality of collaborations
In order to assess the quality of collaborations based on the experiences and assessments of the scientists, they were asked what worked well and what did not work so well. For this purpose, several functional quality dimensions were assessed individually and compiled (Figure 24). Generally, the collaborations functioned mostly “rather well” to “very well”. It should be emphasized that the “strong” conditions for success, namely the "fulfilment of project goals" and the "fulfilment of one’s own goals", were rated as functioning "rather poorly" or "very poorly" only by a very few respondents and as functioning "rather well" or "very well" by the vast majority. It seems to be easy to establish trust between collaborative partners, which can be regarded as a fundamental condition for the success of collaboration.
It is remarkable that the “integration of different professional perspectives (interdisciplinarity)” caused fewer difficulties than perhaps expected. Difficulties were most likely to occur in the “functioning division of labor”, the “fair distribution of funds and resources”, and the “integration of different work styles”.
Figure 24 Quality of collaborations
A comparison of status groups showed that predocs were more critical of collaborations than postdocs in all aspects and that professors gave the best partial ratings. The differences were particularly large in the quality assessments of the “functioning division of labour” and of the “fair distribution of funds and resources” (see Figure 25). These differences reflect the specific roles that predocs, postdocs, and professors play in research processes and, accordingly, in collaborations.
Figure 25 Quality of collaborations by status groups
Status-specific differences also emerged for the central quality dimensions "fulfilment of project goals" and "fulfilment of one’s own goals": for all status groups, "fulfilment of one’s own goals" dropped significantly compared to "fulfilment of project goals". In the case of non-doctoral scientists, however, both values were somewhat lower. A good 83% of professors, but only just under 76% of predocs, stated that their own goals could be achieved in collaborations (see Figure 26).
Figure 26 Fulfilment of goals in collaborations by status groups
The situation is similar when comparing the subject groups: While there were hardly any differences with regard to the fulfilment of project goals, there were differences when it came to achieving one's own goals: Almost 85% of natural scientists rated the achievement of their own goals in collaborations as good. For engineering scientists, this was only 72% (see Figure 27). Thus, it was also in the engineering sciences that the fulfilment of the project goals and the fulfilment of own goals diverged the most. It is reasonable to assume that this was due to the more frequent occurrence of corporate collaborations in the engineering sciences (keyword: contract research), which we were able to show in Figure 9. Engineering scientists collaborated more than other subject groups with companies and were subject to strict requirements at the same time. The difficulty here could be that scientists’ own goals sometimes take a back seat.
Figure 27 Fulfilment of goals in collaborations by subject groups
A comparison of male and female scientists revealed no significant differences (see Figure 28). There were also no major differences when comparing the organizational forms of universities (BUA) and non-university research institutions (BR50, see Figure 29).
Figure 28 Fulfilment of goals in collaborations by gender
Figure 29 Fulfilment of goals in collaborations by organizational forms
Factors influencing the quality of collaborations
The previous analyses were descriptions of bivariate relationships. In order to weigh the influence of different factors on the quality and success of collaborations, multivariate analyses and, in this case, multiple regressions are adequate methods. Figures 30 and 31 identify the factors that have an influence on the quality of collaborations and, thus, their success. The models show that the level of research collaboration affects both success parameters, namely the “fulfilment of project goals” as well as the “fulfilment of one’s own goals”. This means that the higher the level of cooperation, the more likely it is that the success of the cooperation will be ensured. It can be deduced from this that frequent cooperation, or collaboration experience, has a positive effect on the quality of cooperative relationships (see Figures 30 and 31).
Figure 30 Factors influencing the quality of collaborations (fulfilment of one’s own goals)
Figure 31 Factors influencing the quality of collaborations (fulfilment of project goals)