Virtue in Their own Words: AI-assisted analysis of moral identity in student narratives
Faculty Sponsor(s)
Brittany Tausen, Ph.D.
Presentation Type
Event
Primary Department
Psychology
Description
One hundred and thirteen students were analyzed with ChatGPT to rank (1-7) each story’s alignment with the Jubilee Centre’s four virtues categories: intellectual, performance, moral, and civic. There was acceptable interrater reliability (Cronbach’s α’s = .60-.67). Moral virtues were the most prominently expressed (M = 6.11), followed by performance (M = 5.71), intellectual (M = 5.31), and civic virtues (M = 4.93). Bonferroni-corrected comparisons show moral virtues are significantly more prominent than civic, t(112) = 11.66, p < .001, d = 1.10; intellectual, t(112) = 7.51, p < .001, d = 0.71; and performance virtues, t(112) = 3.85, p < .001, d = 0.36.
Copyright Status
http://rightsstatements.org/vocab/InC/1.0/
Additional Rights Information
Copyright held by author(s).
Virtue in Their own Words: AI-assisted analysis of moral identity in student narratives
One hundred and thirteen students were analyzed with ChatGPT to rank (1-7) each story’s alignment with the Jubilee Centre’s four virtues categories: intellectual, performance, moral, and civic. There was acceptable interrater reliability (Cronbach’s α’s = .60-.67). Moral virtues were the most prominently expressed (M = 6.11), followed by performance (M = 5.71), intellectual (M = 5.31), and civic virtues (M = 4.93). Bonferroni-corrected comparisons show moral virtues are significantly more prominent than civic, t(112) = 11.66, p < .001, d = 1.10; intellectual, t(112) = 7.51, p < .001, d = 0.71; and performance virtues, t(112) = 3.85, p < .001, d = 0.36.
Rights Statement
Comments
Submitted to EWU in Bellevue but have not been accepted yet