Mike Bastedo sent this to me; it’s a field experiment looking at how faculty respond to email requests from prospective doctoral students of different races and gender. Their Figure 2 is pretty amazing; I would not have guessed the differential was so large within the field of education (although note the size of the standard error):
