Details about statistical consultation
The statistical consultation of scientific work is a service of the Department of Medical Biometry and Statistics which is offered to employees working with the University Medical Center Freiburg, the theoretical-medical institutes and to assigned academic training hospitals, as well as students graduating from the medical department Freiburg. As far as possible, employees, graduate students and students taking their diploma thesis belonging to other departments can utilise this service, too. The possibility of consultation should be already taken in the planning phase of the scientific work to avoid non-significant results. To have a basis for an effective consultation, it is expected that the examined question and the goal of the investigation are written down in a precise way.
This description has to be provided in the online application form. It is a necessary requirement for the consultation and, if the work is a dissertation, has to be discussed with the mentor. For PhD works, in particular for several coherent works, the mentor should be included in one of the first consultations.
Larger projects, which are to be executed in several dissertations, must be registered for consultation as a single project. The consultation will then be provided for the entire investigation only and not for the analysis of partial aspects.
The statistical analysis itself is usually not conducted by the responsible consultant. There might be special projects demanding the analysis by scientists of the Department of Medical Biometry and Statistics. In this case, the data has to be provided in a form which can be directly read and edited by the department's statistics software. Further details are described in the supplementary sheets "Demands on the file and data structure necessary for evaluation" and "excel2sas". The person seeking consultation is solely responsible for data input and quality. The approximate length of time for consultation should be discussed with the consultant, and especially longer interruptions need to be communicated ahead of time. After completion of the work, the time for which documents as well as data and analysis software should be stored has to be determined in coordination with the consultant.
The consultation, including the use of the computing infrastructure, is free of charge for purely scientific works. If the department contributes substantially to the work and the cooperation results in a publication, the consulting scientist should be included as a co-author. If this is the case, the work has to be submitted to the department in advance of its publication and/or submission.
The consultation and the support for analysis and evaluation assume basic knowledge in statistics, such as it is taught in QB1 Epidemiology, Medical Biometry and Medical Informatics. Recommended literature:
Schumacher M, Schulgen G: Methodik klinischer Studien. Methodische Grundlagen der Planung, Durchführung und Auswertung., New York - Heidelberg: Springer, 3. Auflage , 2008: