Table 5.
Expert Ratings for each Study Case and Combined
| Case 1: Anna | Case 2: Josh | Combined | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quality | Quality Definition | Median | Mean | Standard Deviation | Range | Median | Mean | Standard Deviation | Range | Median | Mean | Standard Deviation | Range |
| Tone | Ability of chatbot to express information in a way that is appropriate for the type of information being delivered. | 4 | 4.25 | 0.71 | 3–5 | 4 | 4.25 | 0.71 | 3–5 | 4 | 4.25 | 0.68 | 3–5 |
| Clarity | Ability of chatbot to communicate information clearly and in a way that avoids ambiguity or confusion. | 4 | 3.88 | 1.1 | 2–5 | 4 | 3.75 | 1.0 | 2–5 | 4 | 3.81 | 1.05 | 2–5 |
| Program Accuracy | Ability of chatbot to provide correct information about the In Our DNA SC program | 3.5 | 3.25 | 1.58 | 1–5 | 3.5 | 3.25 | 1.28 | 1–5 | 3.5 | 3.25 | 1.39 | 1–5 |
| Domain Accuracy | ability of chatbot to provide correct information about the genetic test results and care implications | 4 | 3.88 | 0.83 | 2–5 | 4 | 3.38 | 1.06 | 1–4 | 4 | 3.63 | 0.96 | 1–5 |
| Robustness | Ability to handle ambiguous queries or incomplete information | 4 | 3.75 | 0.71 | 3–5 | 4 | 3.88 | 0.64 | 3–5 | 4 | 3.81 | 0.66 | 3–5 |
| Efficiency | Ability to provide answers that are direct, concise, and complete | 4 | 4 | 1.07 | 3–5 | 3.5 | 3.75 | 1.16 | 2–5 | 3.5 | 3.88 | 1.09 | 2–5 |
| Boundaries | Ability to avoid answering questions that are unrelated to the topic | 4 | 4 | 0.76 | 3–5 | 4 | 4 | 0.76 | 3–5 | 4 | 4 | 0.73 | 3–5 |
| Usability | Ease of interfacing with Chatbot | 4 | 4.38 | 0.52 | 4–5 | 4 | 4.13 | 0.64 | 3–5 | 4 | 4.25 | 0.58 | 3–5 |
| Average Scores | 3.92 | 3.94 | 0.92 | 1–5 | 3.80 | 3.88 | 0.91 | 1–5 | 3.88 | 3.86 | 0.89 | 1–5 | |