Presentation: Ivory tower or transdisciplinarity? Attitudes and preferences of German scientists towards knowledge transfer and co-production

Many German non-university research organizations –in particular institutes of the Max Planck society, Fraunhofer society, Leibniz association, and Helmholtz association– have a long history of engaging in some types of collaborative knowledge creation and transdisciplinary research. However, research disciplines and methodologies, organizational approaches as well as individual mindsets differ widely across and within the four organizations, making it particularly hard to incentivize research teams to tackle current challenges by co-producing actionable knowledge jointly with practice partners from both the public and the private sector. We take a step back and measure the preferences of scientists in these German key research organizations regarding knowledge co-production and more. We conducted a discrete choice experiment accompanied by further survey questions, administered online and distributed via internal communication departments. 917 respondents completed our survey. We asked participants to choose one out of two hypothetical research projects characterized by six attributes (funding volume, academic success, societal impact, knowledge transfer, knowledge co-production, and practice partner type), of which most had three levels (low, medium, high). Respondents were randomly allocated into two blocs, both of which were asked to make eight choices. Standard logistic regressions produce highly significant coefficients for virtually almost all attribute levels. Obvious preference rankings were all confirmed (i.e. researchers prefer high over low academic success, ceteris paribus). Contrasting organization-specific sub-samples reveals clear preferential differences between researchers from different research associations, in particular Max Planck vs Fraunhofer, where the former prefer knowledge transfer to the public sector, while the latter are keener to engage in knowledge co-production with the private sector. However, latent class analysis and other advanced econometric techniques show that there are also intricate differences of preferences within research societies, rendering blanket policies to encourage research-practice collaboration too simplistic. Based on the results of our survey we discuss approaches how to foster transdisciplinarity among the thousands of top scientists in German non-university research.