My personal experience suggests that achievement in research correlates only poorly with both clinical skills and teaching ability. One way of reducing fraud in science is to stop blackmailing people into undertaking research to which they are not intellectually committed and to give to those who are considerably improved facilities.