Profiling Faculty Adoption of Generative AI: A Latent Profile Analysis of Platform Engagement and Inferred Drivers
This study utilizes Latent Profile Analysis on objective platform log data from 358 faculty members to identify two distinct Generative AI usage profiles—"Teaching-Assistance" and "Research-Support"—thereby bridging the gap between self-reported intentions and actual behaviors to inform differentiated institutional support strategies.
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 understand how a group of people uses a new, magical tool. In the world of science, this is often called "technology adoption." For a long time, researchers have asked people, "Do you like this tool? How often do you use it?" But there's a catch: people aren't always honest. They might say they use a tool every day because they think it sounds cool, or they might forget the times they actually used it. This gap between what people say they do and what they actually do is a big problem. To fix this, scientists are starting to look at "digital footprints"—the invisible trails we leave behind when we click, type, and scroll on computers. By studying these footprints instead of asking questions, researchers can see the real story. This paper dives into that world, looking at how university teachers use a new kind of "magic brain" called Generative AI. It asks: Are teachers just using it to make their homework easier, or are they using it to discover new secrets of the universe? And does it matter if they are good with computers or just starting out?
This study, titled "Profiling Faculty Adoption of Generative AI," takes a fresh look at how 358 university teachers in Dalian, China, actually used a smart learning platform between January and June 2025. Instead of sending out surveys and hoping for the truth, the researchers acted like digital detectives. They analyzed the "logs"—the automatic records of every click, every minute spent, and every function used by the teachers. They wanted to see if there were hidden patterns in how these teachers behaved.
Using a special math technique called "Latent Profile Analysis" (think of it as a super-smart way to sort people into groups based on their habits), the researchers found that the teachers naturally fell into two very different camps. They didn't just use the AI a little or a lot; they used it for completely different reasons.
The first group is the "Teaching-Assistance Type." Imagine these teachers as the "Classroom Helpers." They mostly show up on weekdays, right during normal school hours, just like a regular workday. They use the AI to do things that help them teach: generating lesson plans, grading assignments, and creating study materials. These teachers tend to be beginners with the technology. They use the tool to make their daily teaching job smoother and faster, but they don't usually dive deep into complex research tasks. They are the reliable workers who use the tool to get their main job done.
The second group is the "Research-Support Type." These are the "Deep Divers." They are much more active during evenings and weekends, times when most people are relaxing. They don't just use the AI for simple tasks; they use it for heavy lifting like writing academic papers, managing complex knowledge, and analyzing research data. This group is made up mostly of experts who are very comfortable with technology. They treat the AI like a powerful partner in their scientific discoveries, using it for long, sustained periods to solve hard problems.
The study suggests that these two groups are quite distinct. The "Teaching-Assistance" teachers are like commuters who use the train to get to work on time, while the "Research-Support" teachers are like explorers who use the train to travel to new, uncharted lands late at night. The data showed that while both groups use the tool, their habits, timing, and skill levels are very different.
One interesting thing the study found is that being good with computers matters. The "Deep Divers" (Research-Support) were much more likely to be experts, while the "Classroom Helpers" were mostly beginners. This suggests that if you want teachers to use AI for deep research, they might need more training and confidence first.
The researchers are careful to say that this is based on data from just one university, so we can't be 100% sure this is true for every school in the world. However, the patterns they found are clear and strong within their data. They argue that schools shouldn't treat all teachers the same. Instead of giving everyone the same boring training, schools should offer different kinds of help: quick, easy tips for the "Classroom Helpers" who just want to grade papers faster, and advanced, deep-dive workshops for the "Deep Divers" who want to push the boundaries of science.
By looking at the actual digital footprints instead of just listening to what people say, this paper opens a window into the real world of how teachers are adapting to the future. It shows that while everyone is using the same tool, they are using it in two very different ways to solve two very different problems.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.