Table 3 Mean scores of evaluation of GPT-4o GenAI selection of consistent and strongly supportive quotes by specific sentiment.
From: Evaluation of large language models within GenAI in qualitative research
Sentiment | AGYW, mean scores | Male, mean scores | ||
|---|---|---|---|---|
Quotes are consistent1 | Quotes are strongly supportive2 | Quotes are consistent1 | Quotes are strongly supportive2 | |
Anger | 5.7 | 9.3 | 2.7 | 3.7 |
Disgust | 0.33 | 0.33 | 1 | 2 |
Fear | 5.3 | 10.3 | 5 | 10 |
Joy | 6 | 12 | 2 | 2.7 |
Negative | 6 | 12 | 5.7 | 8.3 |
Positive | 6 | 12 | 3.7 | 6.7 |
Sadness | 5.7 | 11 | 0 | 0 |
Surprise | 4.7 | 10 | 3 | 5.3 |
Trust | 5.3 | 10.7 | 2.7 | 5.3 |
