Modeling Artificial Intelligence Tool Adoption in Mathematics Teacher Education: A TAM-Based Structural Equation Modeling
This study utilizes an augmented Technology Acceptance Model to analyze AI adoption intentions among 268 pre-service mathematics teachers in Patna, India, revealing that while Perceived Usefulness and Ease of Use strongly drive positive attitudes and behavioral intentions, external factors like Technological Self-Efficacy and Facilitating Conditions have a relatively low direct impact.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are trying to convince a group of future math teachers to start using a super-smart, AI-powered robot assistant in their classrooms. You might think, "If they just believe they are good with computers, or if the school gives them good Wi-Fi, they will use it!" But a new study from India suggests that's not quite how the magic works.
The researchers, Vivek Kumar Rawat, M. T.V. Nagaraju, and Kumari Dipti, decided to test this idea using a "map" called the Technology Acceptance Model (TAM). They asked 268 student teachers in Patna, India, to fill out a survey about their feelings toward AI tools. Think of this survey as a giant, digital mood ring that measures six different things: how useful they think the AI is, how easy it is to use, how they feel about it, whether they plan to use it, how confident they feel with tech, and what support the school provides.
The Big Surprise: It's Not About Confidence or Support (At Least, Not Directly)
Here is the twist: The study found that simply feeling confident in your tech skills (called "Technological Self-Efficacy") or having a school that promises to help (called "Facilitating Conditions") didn't directly make the students want to use the AI in this specific context.
In fact, the data showed that the link between "I feel confident" and "I think this tool is useful" was statistically non-significant for this group. The researchers noted that these factors showed "little direct effect," indicating that belief in one's own abilities and the support provided by the environment do not carry predictive value for adoption unless their experiential value is made explicit. Even the idea that "if I plan to use it, I will find it easy" didn't hold up in their numbers.
The Real Drivers: "Is it Useful?" and "Is it Easy?"
So, what did work? The study found that the two biggest keys to unlocking the students' desire to use AI were Perceived Usefulness and Perceived Ease of Use.
Think of it like this: Imagine you are holding a new, fancy video game controller.
- Perceived Usefulness (PU): This is the question, "Will this controller actually help me beat the boss level?" The study found that if the student teachers believed the AI would actually make their future math lessons better, they were much more likely to like it and want to use it. This was the strongest factor, acting like a heavy magnet pulling them toward the technology.
- Perceived Ease of Use (PEOU): This is the question, "Is this controller going to be a nightmare to figure out?" If the tool felt easy to handle, it made the students feel better about using it.
The study showed that when students saw the AI as useful, it made the tool feel easier to use, and that combination created a positive attitude. It's like finding out a new app not only saves you time (Useful) but also has a button that says "Do it for me" (Easy). That's when you actually click "Download."
The Numbers Don't Lie
The researchers didn't just guess; they ran the numbers through a complex statistical engine called Structural Equation Modeling. The results were incredibly clear.
- The "Usefulness" factor had a strong, positive effect on "Attitude" (a score of 0.532).
- The "Ease of Use" factor also helped, though a bit less strongly (a score of 0.127).
- The model they built fit the data so perfectly that the error rate was essentially zero (RMSEA = 0.000) and the "fit" score was a perfect 1.000. This means their map of how these feelings connect is very accurate for this group of students.
The "Mediation" Mystery
The researchers also tested a specific theory: Does "feeling confident" (Self-Efficacy) make you think the tool is "useful," which then makes you like it? They checked this path carefully. The result? No. The indirect path was too small to count. In the world of these 268 student teachers, just feeling confident didn't trick them into thinking the AI was useful. They needed to see the actual benefit first.
The Takeaway
So, if you are a school leader or a teacher trainer wanting to bring AI into math education, this study suggests a clear path. Don't just buy the computers and tell the teachers, "You can do it!" or "We have great support!" While those things are nice, they aren't the magic switch on their own.
Instead, you have to show them exactly how the AI will make their math lessons better (Usefulness) and prove that it won't be a headache to learn (Ease of Use). Once they see that the tool is a helpful, easy-to-use partner, their attitude will shift, and they will be ready to adopt it. The study confirms that for these future teachers, the "why" and the "how" matter much more than the "I can" or the "We have," unless those factors are tied to clear, explicit experiences of the technology's value.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.