<p>I am well aware that there is a range of admissions standards at Berkeley, and that engineering, in particular, is extremely competitive. So it may well be that the figures for the engineering students admissions profiles would look different from the university as a whole. Also possible that the grade distributions would be different, and that the relationship between SAT and GPA would differ. Unfortunately, we come back to data. If anyone has this data for the engineering, or other schools, it would be great to see. Until we have it, we are back to speculation.</p>
<p>At a place like Princeton, the assumption is that essentially anyone who gets in can do well enough to go to medical school, or equally competitive graduate or professional school if science is not their thing. The mean GPA of Princeton grads who go to med school is about the same as the overall mean. So “average Princeton student”=“successful med school applicant”.</p>
<p>The only data I could find on GPAs for Berkeley did not break down by major or even STEM vs other. It is certainly possible that the culture tells people that only top students attempt the STEM fields, and that Berkeley STEM graduates look much like Princeton STEM grads. Of course, many premeds do not major in science, and take most of their courses outside of STEM. Again, if anyone know of data on this, it would be great to see.</p>
<p>Princeton has engineering students too, and they probably have a higher academic profile than even the lofty mean of Princeton students overall. So we would need to compare Berkeley engineers to Princeton engineers. Great if we could do that. Do you know of a data source?</p>
<p>Meanwhile to the question of whether medical schools adjust their interpretation of GPAs based on the undergraduate college.</p>
<p>There is data that suggests that perhaps they should, a little. In a broad review of the predictive ability of MCAT, undergraduate GPA, and selectivity of undergraduate institution, the authors found that uGPA alone was a good predictor of medical school grades and USMLE scores. By taking into account the selectivity of the undergraduate college (mean SAT), they improved the prediction by a small but significant amount. This implies that, at a national level, an equivalent GPA from a more selective vs less selective college implies a stronger student. At this point it would suggest that the med school should give a few points for getting equal GPAs from a more selective vs a less selective college. </p>
<p>On the other hand, they found that the MCAT was a substantially better predictor than uGPA, and that with the MCAT score in the model the selectivity no longer contributed to prediction. So whatever effect the selectivity of the undergraduate institution may have, it is fully reflected in MCAT scores.</p>
<p>Note: this means that medical school need not consider the selectivity of the undergraduate college in predicting medical school grades. However it does not tell us whether medical schools actually consider undergraduate college when making decisions.</p>
<p>[Validity</a> of the Medical College Admission Test for Predictin… : Academic Medicine](<a href=“Validity of the Medical College Admission Test for Predictin... : Academic Medicine”>Validity of the Medical College Admission Test for Predictin... : Academic Medicine)</p>