Regarding the main questions of the study that have been formulated above, two findings seem to be most important: Firstly, our study has shown, that it is in fact practically possible to implement PBL as an educational strategy under the conditions of a large German medical school. Secondly, the study provided sound evidence that the PBL approach does not lead to disadvantages concerning students' factual knowledge as measured by a combination of multiple-choice and short-essay questions. The results of our study therefore support earlier findings2[3]–4 and constitute another link in the chain of evidence indicating that PBL is equipotent to LBL as far as students' factual knowledge is concerned.Further conclusions are more difficult to draw: The fact that the PBL students achieved slightly better results in the category of short-essay questions might indicate, that an examination testing more complex levels of knowledge (i.e. on the comprehension or analysis level), is more appropriate for testing the outcome of PBL in terms of factual knowledge than multiple-choice questions. An intra-group comparison of the two categories of questions further supported this hypothesis, revealing that students who had undergone the lecture-based course scored significantly lower in the short-essay part, whereas their PBL counterparts reached similar results in both categories of questions. It is important to stress though, that the direct comparison of the short-essay results in both groups only showed marginal results. Also, the multiple-choice questions we used in our study were not intended to assess a more complex, i.e. comprehension or analysis level of knowledge. Hence, if there is a difference, it is not necessarily a difference between multiple-choice and short-essay questions, but between questions addressing different levels of knowledge.