This is a question for researchers in the field of science education: has anyone ever looked at the correlation between
1) the average student evaluation score of instructors of lower division undergraduate physics for pre-med students
2) the score of these students on the MCAT, which has a very important physics component (as this is a standardized test, this seems to be more cogent an indicator than the student’s grade in the physics class taught by the instructor).
If the correlation were found to be weak or non-existent (hypothetically speaking), then would one be entitled to ask what exactly student evaluations are measuring ? Would one be entitled to posit that maybe student evaluations do not really measure one’s effectiveness at teaching, as much as at making students feel good about themselves ?