Skip to main content

Table 3 Scientific accuracy of rationale for correct and incorrect responses

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

Answers (N)

Rationale (%)

Scientifically accurate

Scientifically inaccurate

Overall

p-value

Correct (178)

73.5

2.6

76.1

p < 0.05*

Incorrect (56)

4.3

19.6

23.9

Subtotal (234)

77.8

22.2

100.0

  1. *The Fisher’s exact test and the follow-up chi-squared test (Χ2(df = 1, N = 234) = 152.93) found significant interaction between the scientific accuracy of rationale and the accuracy of answers to the MCQs (p < 0.05)
Back to article page

BMC Medical Education

ISSN: 1472-6920

Contact us