Statistical projections suggest that by 2020 the number of general practitioners (GPs) in Germany will have decreased by about 13%, which will lead to a risk of medical undersupply, especially in rural areas. This study was conducted on behalf of the Bavarian Ministry of Environment and Health in order to answer the following questions: How satisfied are young Bavarian physicians in GP specialty training with their training situation? To what extent are they willing to start a private practice, in general and, in particular, in rural areas? The study's online questionnaire contained items concerning the context of GP specialty training and the willingness to set up a private practice. In addition, the "Heidelberger Niederlassungsbarometer", a structured questionnaire illustrating self-assessment of competencies of GPs in training, was included in the survey. We asked GP specialty trainees to participate by e-maiI, directly or via their GP trainers. In total, 99 questionnaires were returned, whereof 95 fulfilled the criteria for analysis. In 76% of the cases participants were female, 63% working in practice and 37% in a structured training programme. The majority of participants (58%) preferred to work in a practice with more than one GP, followed by those preferring to work as a salaried GP (26%). Their disposition to work in rural areas after having passed their specialty certificate examination was high or very high (59%). Feedback on a regular basis was given to 29% of the study participants. Almost every second participant received a structured initial training programme. Most of the participating GP trainees (59%) were satisfied or very satisfied with their training. In general, there is a high level of motivation among study participants to become an independent and self-employed GP, also in rural areas. Despite the fact that not even every second received central training elements like individual feedback and structured initial training, the participants were mostly satisfied with their GP specialty training.