Table 3. Using a one-way ANOVA and Tukey post-hoc analysis, we found no significant difference in ChatGPT-calculated DISCERN scores between academic/news and private clinic sources.
* The mean difference is significant at the 0.05 level.
| Reviewer and Article Type | Reviewer and Article Type | Mean Difference | Std. Error | p-value | 95% Confidence Interval | |
| Lower Bound | Upper Bound | |||||
| Human Academic | Human Private | 1.17000* | 0.21542 | <0.001 | 0.5975 | 1.7425 |
| ChatGPT Academic | 0.12118 | 0.22707 | 0.95 | -0.4823 | 0.7246 | |
| ChatGPT Private | .57356* | 0.21542 | 0.049 | 0.0011 | 1.146 | |
| Human Private | Human Academic | -1.17000* | 0.21542 | <0.001 | -1.7425 | -0.5975 |
| ChatGPT Academic | -1.04882* | 0.21542 | <0.001 | -1.6213 | -0.4763 | |
| ChatGPT Private | -.59644* | 0.2031 | 0.025 | -1.1362 | -0.0567 | |
| ChatGPT Academic | Human Academic | -0.12118 | 0.22707 | 0.95 | -0.7246 | 0.4823 |
| Human Private | 1.04882* | 0.21542 | <0.001 | 0.4763 | 1.6213 | |
| ChatGPT Private | 0.45238 | 0.21542 | 0.167 | -0.1201 | 1.0249 | |
| ChatGPT Private | Human Academic | -.57356* | 0.21542 | 0.049 | -1.146 | -0.0011 |
| Human Private | .59644* | 0.2031 | 0.025 | 0.0567 | 1.1362 | |
| ChatGPT Academic | -0.45237 | 0.21542 | 0.167 | -1.0249 | 0.1201 | |