Skip to main content
. 2023 Nov 13;23:864. doi: 10.1186/s12909-023-04832-x

Table 6.

The credibility of provided paragraphs between the correct and incorrect answers

Answers to the MCQs (N) Correct and relevant Paragraph % Correct but irrelevant Paragraph % Incorrect Paragraph % Overall % P-value
Correct (49) 75.4 0.0 4.9 80.3 p < 0.05*
Incorrect (12) 1.6 9.8 8.2 19.7
Subtotal (61) 77.0 9.8 13.1 100.0

*The Freeman-Halton extension of Fisher’s exact test found significant interaction between the correctness of ChatGPT’s response and the credibility of paragraphs from the references (p < 0.05)