In this paper the authors review the few publications on this issue: assessment in interpreting, pointing out the different teaching philosophies that inevitably condition the assessment system applied. There would seem to be some difference of opinion regarding the point at which assessment should be most selective: at the start of an interpreting course through stringent entrance testing, or at the end. There are also differences regarding assessment criteria: the intuitive method (Longley, 1989; Dubrovsky and Welter, 1990) versus the 'discrete components approach' (giving marks for different aspects of the interpretation performance) (Schjoldager, 1996). The authors analyze the different aspects of these issues and put forward their own model. This model aims at combining the parameter most widely used to assess quality in the professional world -intratextual coherence- with parameters that measure the extent to which teaching objectives have been achieved, including intertextual coherence.