The assessment of clinical competence has traditionally been carried out through standard evaluations such as multiple choice question and bedside oral examinations. The attributes which constitute clinical competence are multidimensional, and we have modified the objective structured clinical examination (OSCE) to measure these various competencies. We have evaluated the validity and reliability of the OSCE in a paediatric clinical clerkship. We divided the examination into the four components of competence (clinical skills, problem-solving, knowledge, and patient management) and evaluated the performance of 77 fourth-year medical students. The skill and content domains of the OSCE were carefully defined, agreed upon, sampled and reproduced. This qualitative evaluation of the examination was both adequate and appropriate. We achieved both acceptable interstation and intertask reliability. When correlated with concurrent methods of evaluation we found the OSCE to be an accurate measure of paediatric knowledge and patient management skills. The OSCE did not correlate, however, with traditional measures of clinical skills including history-taking and physical examination. Our OSCE, as outlined, offers an objective means of identifying weaknesses and strengths in specific areas of clinical competence and is therefore an important addition to the traditional tools of evaluation.