Skip to main content
. 2026 Mar 3;26:943. doi: 10.1186/s12903-026-07996-2

Table 3.

Model performance by question type

ChatGPT Gemini
Mean ± SD Median(IQR) Mean ± SD Median(IQR) Test statistics p (Wilcoxon test)
Section
 MCQ 3.89 ± 0.97 4(2) 4,62 ± 0,67 5(1) -3.892 < 0.001
 T/F 4.04 ± 0.79 4(1.250) 4,7 ± 0,54 5(1) -4.056 < 0.001
 Open-ended 4.31 ± 0.65 4(1) 4,62 ± 0,53 5(1) -2.435 0.007
Test statistics 7.92 1.04
p (Kruskal–Wallis Test) 0.019