Table 3. Validity assessment of explanations generated by chatbots.

-			Average CVI	ICC (95% CI)
ChatGPT-4o	TR	First	0.95	0.849 (0.756-0.901)
		Second	0.96	0.850 (0.774-0.897)
		Final	0.97	0.834 (0.753-0.885)
	EN	First	0.96	0.951 (0.936-0.963)
		Second	0.96	0.942 (0.924-0.956)
		Final	0.98	0.885 (0.850-0.912)
Gemini 1.5 Pro	TR	First	0.93	0.862 (0.771-0.911)
		Second	0.94	0.878 (0.820-0.915)
		Final	0.95	0.850 (0.757-0.902)
	EN	First	0.92	0.927 (0.893-0.949)
		Second	0.93	0.925 (0.890-0.947)
		Final	0.93	0.918 (0.877-0.944)

TR: Turkish, EN: English, CVI: Content validity index, ICC: Intraclass correlation coefficient, CI: Confidence interval