Humboldt-Universität zu Berlin - English

In order to find out whether the different attitudes or positionings are dependent on certain influencing factors, we estimated multivariate logistic models. This is used to examine the extent to which attitudes towards autonomy (Fig. 28), the role of scientists in debates (Fig. 29, 30) and views on scientific knowledge (Fig. 31, 32) can be traced back to membership of a particular subject group or status group, gender or characteristics of the research context. Figures 28 to 32 show the results.

With regard to the question of how autonomous science should be in relation to societal demands, there are several strong effects (see Figure 28). For example, professors are more in favor of the autonomy of science, while younger scientists are more inclined to the position that science should serve society.

Women are also slightly more likely to affirm the service function of science towards society. This is an important finding, as the model already controls for variables that are confounded with gender, such as subject group and status. It is therefore a genuine gender effect.

With regard to the subjects, the multivariate model shows slight changes compared to the bivariate findings: Life scientists and engineers are significantly more likely to see science as serving society compared to social scientists. The effects of the subject groups are less strong in this model compared to the bivariate correlations, as epistemic conditions of research are also considered here, which explain part of the subject variance (see Figure 28).

Figure 28 Factors influencing attitudes: Autonomy

Three complementary findings can be found for the epistemic conditions: Scientists who work more theoretically, but also those who are dependent on technical infrastructures, are more in favor of the autonomy of science than scientists from other research contexts. Respondents who work in research contexts with high competition from other research groups tend to favor the position that science should serve society (see Figure 28). 

Figure 29 illustrates that only one epistemic characteristic has a significant influence on the position of scientists' involvement in public debates: those who essentially work theoretically / conceptually tend more towards the position that scientists should stay out of public debates.

Figure 29 Factors influencing attitudes: Engagement in public debates

There is also a significant gender effect with regard to the question of whether scientists should only comment publicly on their own research topics or also on other topics (see Figure 30). Here, it is women who are significantly more likely to think that scientists should limit themselves to statements on their own research topics. In addition, scientists from the humanities, life sciences and natural sciences are more inclined to comment on topics that go beyond their own research topics compared to those from the social sciences.

Figure 30 Factors influencing attitudes: Restriction of statements

Figure 31 shows that the attitude towards the value freedom of knowledge cannot be significantly explained by any of the characteristics tested. This means that the variance here cannot be traced back to structural or contextual characteristics and therefore tends to reflect subjective, individual positioning.

Figure 31 Factors influencing attitudes: Axiological neutrality

Figure 32 illustrates that life scientists, natural scientists and engineers agree significantly more often with the position that knowledge is objective and universally valid compared to social scientists. In contrast, scientists whose research is characterized by strong competition with other research groups are more likely to agree with the position that knowledge is always open to interpretation and provisional (see Figure 32).

Figure 32 Factors influencing attitudes: Objectivity