Effects of Entry Grades on Students' Academic Performance Under Homogeneous Educational Resources

A minor difference in college entrance examination scores can result in vastly different educational resources in China, so it has been debated whether it is the difference in the student population or the difference in educational resources that causes the difference in medical graduates. We aimed to evaluate the effects of entry grades on students' academic performance under homogeneous educational resources.

Students in grade 2016 with 13-point difference in the average admission scores of 2 medicine schools in Sun Yat-sen University were educated in mixed classes and were taught with the same educators during the 5 years of undergraduate period. The grades, graduation, and postgraduate enrollment rates of the students were compared between the two campuses.

The average admission scores for Shenzhen Campus (SZC) students are 13 points lower than those of Guangzhou North Campus (GZNC) (613 points vs 626 points). After 5 years of homogeneous education, comparing the GZNC students with the SZC students, there were no significant differences in the average total score (80.2 ± 4.6 vs 80.0 ± 5.6, P = 0.691), the average compulsory course (78.9 ± 3.4 vs 78.4 ± 6.1, P = 0.438), the average core course score (78.8 ± 7.4 vs 78.7 ± 5.0, P=0.860) and the average clerkship score (85.1 ± 7.2 vs 84.6 ± 2.7, P=0.275). However, the completion rate for SZC was higher than for GZNC (93.94% vs 86.27%, P=0.009). There was no statistical difference in postgraduate enrolment between the two institutions (P=0.758).

Given the same educational resources, more medical students with lower entrance scores completed their studies and achieved the same percentage of postgraduate acceptance. This finding suggests that a key component of improving the quality of medical higher education in China may be to further rationalize the allocation of high-quality educational resources, rather than to pursuing students with high entrance examination scores.