Heads of universities and research institutes are responsible for an adequate organizational structure. Taking into account the size of each scientific unit, the responsibilities for direction, supervision, conflict resolution, and quality assurance must be clearly allocated, and their effective fulfillment must be verifiable.

In science as in all other fields, adherence to fundamental values is particular to each individual. Every scientist and scholar is personally responsible for his or her own conduct. But whoever is responsible for directing a unit also carries responsibility for the conditions therein.

Members of a working group must be able to rely on each other. Mutual trust is the basis for the conversations, discussions, and even confrontations which are characteristic of groups that are dynamic and productive. A researcher's working group is not only his or her institutional home base; it is also the place where, in conversations, ideas become hypotheses and theories, where individual, surprising findings are interpreted and brought into a context.

Cooperation in scientific working groups must allow the findings, made in specialized division of labor, to be communicated, subjected to reciprocal criticism and integrated into a common level of knowledge and experience. This is also of vital importance to the training of graduate students in the group for independent research. In larger groups, some organized form for this process (e.g. regular seminars) is to be recommended. The same holds true for the reciprocal verification of new findings. The primary test of a scientific discovery is its reproducibility. The more surprising, but also the more welcome (in the sense of confirming a cherished hypothesis) a finding is held to be, the more important independent replication within the group becomes, prior to communicating it to others outside the group. Careful quality assurance is essential to scientific honesty.

The organization of working groups does not have to be hierarchical. But whether or not this is the case, there will always be a functional division of responsibilities, e.g. when one member of the group assumes the role of principal investigator of a grant proposal, and thereby becomes accountable to the funding institution according to its rules. Usually, one person heads a working group. He or she bears the responsibility that the group as a whole is able to fulfill its tasks, that the necessary cooperation and coordination are effective and that all members of the group are aware of their rights and their responsibilities. This has immediate consequences for the optimum and maximum size of a group. A leadership function becomes void when it cannot be exercised responsibly on the basis of the knowledge of all relevant circumstances. Leading a working group demands presence and awareness. Where – for instance at the level of the direction of large institutes or clinics – these are no longer sufficiently assured, leadership tasks must be delegated. This will not necessarily lead to complex hierarchical structures. The 'leadership chain' must not become too long. Institutions of science are under obligation to provide organizational structures which should ideally promote, but at least permit the type of healthy communication described above. Universities, as corporate institutions, and independent research institutes by analogy, must guarantee working conditions that allow all their members to observe the norms of good scientific practice. Heads of institutions carry the responsibility to ensure that a suitable organizational structure is (and is known to be) in place, that goals and objectives will be set and progress towards them can be monitored, and finally, that mechanisms for resolving conflicts are available.

–> 4.3 Education in Good Scientific Practice