Table 4 Comparing ChatGPT’s performance accuracy between SBAs and EMQs

From: Exploring the capabilities of ChatGPT in women’s health: obstetrics and gynaecology

Question Type

Correct

Incorrect

Total

Single best answer (SBA)

318 (54.0%)

271 (46.0%)

589

Extended matching questions (EMQ)

180 (45.0%)

220 (55.0%)

400

Total

498

491

989

  1. ChatGPT performed better in single best answer (SBA) questions than extended matching questions (EMQ), p = 0.01, χ² = 7.35. Values in brackets denote the percentage proportion (%).
Back to article page