[Docimological analysis of the anatomy test for the PCEMII final exam in Dakar]

This docimological analysis is based on computer treatment of the results of the anatomy exam for the two sessions of July (159 candidates) and October (56 candidates) of the identification questions, discrimination questions, coupling (paining), grouping, association and diagram labelling. The results were appraised following certain parameters: the gross score revealed a rather low average score for the whole class (14.57 and 9.57 over 30): the low success rate in July and the average figure of October indicate, in general, either the difficulty of the questions for the students or a bad reception of the message; the bad discrimination index, since only 21% of the questions are of the discriminatory type. This finding pointed to a revision of 50 of the 68 questions of July and 53 of the 68 questions of October. On the whole, only 15 questions, over the two sessions, could have been kept unchanged for an eventual bank (stock of data): The correlation coefficient has showed a significant relation between success in the anatomy subject and success in the PCEM2 exam. This work has showed the possibilities that computers may offer in the teaching of anatomy and the evaluation of students. Besides, it emphasizes the importance of the anatomy grade in the PCEM2 exam. Finally, the mode evaluation used help exploring the various levels of the cognitive areas (memorization, interpretation of data, problem-solving) and to build up a bank of anatomy questions.