Complex cognitive processes, such as interpretation of information and problem-solving, form a critical part of medical practice; therefore, the aims of many educational programs include teaching these processes. However, systematic evaluation of students' interpretation of information or problem-solving may be difficult. To determine whether objective test items could measure cognitive processes more complex than recall of isolated facts, the committee of the American Academy of Orthopaedic Surgeons preparing the Orthopaedic In-Training Examination classified test items into three taxonomic levels according to the cognitive process required to answer the item: recall or recognition, interpretation, and problem-solving. Orthopedic residents from 20 universities took the examination and classified items according to the cognitive process they believed they used to answer the questions. Two methods of analysis showed that residents agreed closely with the taxonomic levels assigned by the committee, suggesting that items did test the cognitive process they were designed to measure and that objective assessment of complex cognitive processes is practical and potentially very useful in medical education.