Skip to main content

Table 5 The distribution of paragraphs between the correct and incorrect answers

From: Assessment of the capacity of ChatGPT as a self-learning tool in medical pharmacology: a study using MCQs

Answers to the MCQs (N)

MCQs provided with Paragraphs (%)

MCQs not provided with Paragraphs (%)

Overall (%)

P-value

Correct (56)

46.2

25.6

71.8

p > 0.05*

Incorrect (22)

14.1

14.1

28.2

Subtotal (78)

60.3

39.7

100.0

  1. *Fisher’s exact test and a follow-up chi-squared test (χ2(1, 78) = 1.35) found no significant interaction between the correctness of ChatGPT’s response and whether or not it was provided with relevant paragraphs from the references (p > 0.05)
Back to article page

BMC Medical Education

ISSN: 1472-6920

Contact us