•  
  •  
 

Author ORCID Identifier

Oleksii Shapoval: https://orcid.org/0000-0003-4478-3193

Abstract

Knowledge assessment is an integral part of learning and teaching that requires considerable amounts of instructors’ time and effort. This study aims to evaluate the prospects of artificial intelligence (AI) for automated knowledge assessment and identify the factors that influence its perception and value. An experiment with a reference implementation of a grading and automated feedback generation system is performed. Based on the results, a survey with 16 university students who regularly use AI is conducted and analyzed through a thematic analysis method. The findings reveal that availability, knowledge breadth, and elimination of social barriers have positive influence on preference towards AI feedback while specificity, actionability, and encouragement positively influence its perceived usefulness. Excessive length, lack of contextual awareness, and impersonal responses were among the main negative factors. Despite readiness to receive AI feedback, respondents have expressed preference towards feedback from a teacher and revealed the preferred use of AI tools for productivity-enhancing and clerical tasks. The findings confirm that the role of AI in knowledge assessment is in complementing the process rather than substituting the teacher’s role in it.

First Page

169

Last Page

196

Ethics Approval

Yes

Declaration Statement

Data Availability Statement: The data presented in this study are available on request from the author.

Funding statement: No funding has been received in this work.

Conflicts of Interest: The author declares no conflict of interest with respect to this work.

Participation consent statement: Informed consent was obtained from all participants of the study.

Permission to reproduce materials from other sources: N/A

Clinical trial registration: N/A

Download

Share

COinS