We have had some in-depth discussions about the proposal for a structured evaluation at the end of the first year of a PhD program and other related topics. These are aimed at creating a more effective and supportive journey for our doctoral students. We acknowledge that doctoral studies are inherently highly individual. However, we feel that by clarifying crucial milestones a PhD candidate has to achieve, it provides clarity to students and supervisors increasing their focus on research and by motivating students to produce visible output early on in the process. Here is a breakdown of the core aspects we are considering:


Proposal review: We agree with the DocSchool Computer Science on a formal evaluation after the first year to assess whether the proposed PhD project satisfies the quality standards in the field and has sufficient likelyhood of success. This aims to reduce dropouts after several years. While we generally consider a year a good point of reference, the exact duration for developing the PhD proposal can be more flexible and depend on the field and different modes of employment, such as project assistants, university assistants and non-employed doctoral students. We do not consider this a major barrier for the high percentage of doctoral students already performing very well.


Supportive measures for students: If the proposal review suggests that a student's project might not succeed, comprehensive support must be provided by the DocSchool coordination team. This includes help in finding new research topics or supervisors, ensuring students aren't left without options and in a legal vacuum (employment and admission, but no project). As such, we advocate to promote a culture change, where transferring the supervision of PhD students is not considered problematic, nor a personal or academic affront. See also our comment in Section “Assessment pre-admission”.


Co-supervision on an institutional level: We are advocating for the introduction of a thesis committee and co-supervision from the start. This would enable timely and multi-faceted feedback, allowing PhD students to make necessary adjustments early on. Unite! and other initiatives provide a suitable breeding ground for more co-tutelle agreements which we expect to have similar outcomes. In line with national mobility goals, the PhD Union envisions at least 20% international PhD studies by 2026. Co-supervision need not be restricted to people with a venia docendi, but can also be done by PostDocs after suitable in-house training.


Assessment pre-admission: We agree that assessing a student's suitability for a PhD should happen before the admission process, for example using an interview with more involved TU staff. This is crucial to ensure students are fully prepared and understand what the PhD program entails.


Regular progress presentations: While there are certainly a lot of best practices of freqeuent progress presentations to members of the institute and group members, the curriculum should highlight the importance. A particular example can be found in the DocSchool for Molecular Biosciences and Biotechnology where there is an institutionalized thesis committee making decisions in this step, similar to the plans of the DocSchool for Computer Science.


Transitional arrangements: Similar to Bachelor and Master programmes, changes to the DocSchool statutes can have a negative impact on the study progress of doctoral students. We strongly suggest a transition period for statutes and would like to put this into discussion in the curricula comission for doctoral studies and studies for continuing education.


To summarize our points, we strongly suggest measures aimed at providing early feedback for PhDs in order to prevent unnecessary and unsuccessful work for many years both on the student’s as well as on the supervisor’s side. Despite the benefits, we're aware of the concerns regarding the practical, legal, and contractual implications, especially from the workers' council's perspective. It is crucial that we find a balance between academic rigor and these realities.


Note: An early version of the statement has been generated by GPT-4 based on our discussion in writing. More than 90% were completely rewritten, the entire docuiment majorly overhauled and approved by members of the PhD Union.