Imagine a student who works on a task and they are stuck. Their lecturer or lecturer is not available. Or maybe if they ask for help they look stupid. So they turn to ChatGpt instead of feedback.
At the same time, they will have an answer that they can make more explanations if needed.
They are not alone. Our research shows that almost half of the students studied by the University of Australia use artificial intelligence (AI) for feedback.
Our study
Between August and October 2024, my colleagues examined 6,960 students at four major Australia universities.
Participants studied a wide range of topics including science, engineering and mathematics, health, humanities, trade and law.
More than half (57 %) were female, 72 % between 18 and 24 years old. Nearly 90 % of full -time students, 58 % of domestic students, 61 % undergraduate and 92 % as part of their education participated in academic activities.
Each of the universities invited their enrolled students to conduct an online survey.
We wanted to understand how students use artificial intelligence to learn, especially if they used artificial intelligence for feedback, and their understanding of the help and confidence of feedback from artificial intelligence and teachers.
Students think artificial intelligence feedback is useful but unreliable
We found that almost half of the scrutiny (49 %) use AI for feedback to help them improve their university work. For example, this can include typing questions in popular tools such as ChatGpt. It can also include suggestions to improve a workpiece, details of the strengths and weaknesses of the work, the proposed text editing and additional ideas.
The students told us that they considered both artificial intelligence and teachers’ feedback to be useful: 84 % of artificial intelligence feedback as useful, while 82 % said the same thing about their masters.
But when he trusted him, there was a big gap. About 90 % of students find their teacher’s feedback reliable, compared to 60 % for artificial intelligence feedback.
As one of the students said. “[AI] Immediate access to information, descriptions and creative ideas, which can be useful for quickly solving the problem and exploring new concepts. “
Another student said teachers’ feedback is “more challenging but rewarding”. The reason was: “[AI] It seems to confirm some of my thoughts, and that makes me useful for how useful. “
Artificial intelligence provides volume, and teachers have specialized
Our subject analysis of students’ open answers shows that artificial intelligence and teachers offer different goals.
Students reported that artificial intelligence was less reliable and less specific. They also pointed out that artificial intelligence did not understand the field of assignment as well as their teachers.
However, it was easier to access artificial intelligence – students can ask for feedback several times without feeling load.
Vulnerability factor
Research tells us that students can feel vulnerable when searching for feedback from teachers. If their work is not sufficient, they may be concerned about judging, feeling embarrassed, or damaging their academic relationships.
Artificial intelligence seems to eliminate this concern. A student explained how “[AI] Feedback has less security and judgment “. Another student explained:”[AI] Let me ask stupid questions that I am so ashamed to ask your teacher. “
But many students do not know artificial intelligence can help
Half of the participants (50.3 %) did not use AI for feedback purposes – 28 % of this group simply didn’t know that it might.
Other reasons include distrust of artificial intelligence (28 %) and personal values that opposed the use of this type of technology (23 %).
This can create a matter of justice. Students who are aware of the capabilities of artificial intelligence have 24/7 access to some possibly useful feedback support, while others have none.
What does this mean unis
As student participants said, artificial intelligence can be useful in providing quick and accessible feedback for early drafts.
Teachers are superior to providing specialized and textual guidance that gives a deeper understanding. This allows you to get medical advice from a qualified physician against symptoms on Google. Both may be useful, but in different situations, and you know which one you will trust with one serious thing.
For those universities trying to find a way to combine artificial intelligence in their teaching and learning systems, it is a challenge to create opportunities and structures that enable educators to focus on their strengths. Artificial intelligence can complete them by providing useful and digestible information about student work that is easy to understand and is almost always available and free of personal judgment.
This shows that the future is not about the choice between artificial intelligence and human. This is about understanding how they collaborate to support students more effectively.
In this article, Michael Henderson of Monash University, Margaret Barman and Jennifer Chang of Deakin University, Fonmes from Monash University, Simon Buckingham from Sydney University of Technology, Kelly E. Matthews from the University of Queensland, and Jimna de Hordia from Monash University.Human
Jimna is a doctoral candidate at the Faculty of Education at Monash University.
This article is published by Creative Commons. Read Main articleHuman
