Aim. To assess the impact on clinical internship of problem-based learning (PBL) methodology applied to nursing studies. Subjects and methods. Longitudinal and prospective study of two cohorts of nursing students: the intervention group was the promotion 2005-2006 of the nursing school, that received their education with PBL, and the control group was the promotion 2004-2005 of the nursing school, that received their education with other pedagogic methods, mainly lectures and seminars. We analysed the possible correlation in each of the items on the mentor's assessment form between the grade obtained by students and the learning methodology employed during their education (PBL or non-PBL). The assessment criteria were: responsibility; initiative; skill and proficiency in techniques related to manual procedures; organization and methodology in activities carried out; communication/relationship with patients and families; communication/relationship with the work team; and degree to which training objectives were reached. Results. We had 651 valid assessment forms, 383 for the control group and 268 for the intervention group. Clear significant differences were found only in the third year, in the 'skill and proficiency in techniques related to manual process' category (p = 0.007). The control group had higher scores. Conclusions. The results suggest that the use of PBL in students' academic coursework does not modify the score obtained in their clinical internship on the items specified in our assessment tool.