
Jon W. answered 05/04/19
USMLE/Med School Tutor
The USMLE first computes a raw score based on percentage correct. Any items not answered are graded as incorrect. Pilot (experimental) test items are not counted in the score. Based on the raw score, the USMLE then uses a psychometric process to apply a difficulty modifier based on how difficult the questions on the exam were and then converts to a 3-digit score. The difficulty of each item can be determined when the test item is piloted based on the percentage of test-takers who answer the question correctly.
"All scores are made comparable through equating, a psychometric process that adjusts scores based on the difficulty of the questions. This can be thought of as small score increases applied to examinees who see somewhat more difficult sets of test questions, and small score decreases applied to test takers who see somewhat less difficult sets of test questions."