Authored by Lingxi Zhang
There can be no ignoring the emergence of generative AI technologies. They are starting to have a significant influence on how we write, how we learn, how we search, and in many ways, even what we create. In classrooms and libraries, in group chats and study rooms, AI is now a constant presence. Sometimes the reception is generally positive, other times AI is met with skepticism, however, regardless, AI is quickly becoming embedded in most contexts.
For all of its life-altering potential of AI, particularly in education, it has never really felt like a tech revolution to me; in fact, it has always seemed like a human one. And the narrative that drove me to undertake this work didn’t start with a dataset; it started with a friend.
The Friend Who Inspired the Research Question
This friend is among the most capable people I know—smart, curious, driven by a desire to be better. But her sweet little nesting place for learning has always been like a delicate popsicle, melting just when she needs it most. Even when she enjoys her major, even when her objectives are clear, it is often a challenge for her to maintain the willpower required to complete assignments or remain engaged.
I watched her go from frustration to guilt to redoubled effort and back again, then grow exhausted. And I found myself wondering: Why is it so hard for some students, even when they care deeply about a subject, to keep going? What interrupts motivation? What supports it? What can make learning feel impossible—or possible once more?
Out of these questions I arrived at my questions concerning AI’s role in the field of learning motivation research.
About Motivation: What It Is and Why It Matters
In educational psychology, motivation isn’t a single emotion; it’s a constellation of forces that determine whether we’re engaged or not. There are several factors researchers typically mention:
- Self-efficacy: the extent to which students believe that they are capable of accomplishing the task
- Importance: how much they see the task as important (or not)
- Expectation: whether they expect their effort will lead to success.
- Anxiety & cognitive load: does emotional strain make learning harder?
- Autonomy: to what extent students experience control over their learning
Motivation not only impacts the work but also has effects on engagement, dedication, and confidence. This can have compounding effects for future performance in learning environments. In many empirical studies, low self-efficacy has also consistently been predicted to result in lower academic outcomes, higher dropout risk, reduced class participation, and avoidance of difficult tasks.
As AI continues to be discussed as a tool and a concern in educational space, considering the way that AI may impact motivation is a key part of that conversation. As a society, we are at the very beginning of understanding how AI may impact motivation and learning. However, there are a number of questions I started asking about the relationship between AI and motivation:
- Can it diminish feelings of ownership and participation?
- CanAI inspire students to persevere through adversity?
- Can it make learning seem easier and thus more attainable?
Where My Project Starts
The first step, before I even began poking at any databases, was pinning down the heart of my project: I’m curious about why students would or wouldn’t want to use AI and how it’s related to motivation and learning outcomes. I am not interested in whether AI is good or bad in its own right, but instead interested in looking at how it will interact with motivation, which can be fragile.Perhaps AI can help students like my friend by lowering barriers, reducing anxiety, or supporting them as they regain confidence.
In order to investigate this, I used a synthetic data set available in Kaggle with the following types of variables: AI usage frequency, student motivation, academic performance, and demographic or contextual factors. Using synthetic data enables me to go through the entire process without the ethical and privacy considerations of using real student data. The purpose is not to make definitive statements but rather to understand how to ask questions more effectively with data.
How I Envision the Project Playing Out
As I have designed my project, it will follow the following process:
- Clarify conceptual relationships: Map how AI features (feedback, scaffolding, and other task simplification) can interface with aspects of motivation (self-efficacy, anxiety reduction, value).
- Clean and prepare the data: This will include searching for missing values, re-coding variables, searching for outliers, and checking for inconsistencies.
- Conduct exploratory data analysis: Seek cognitive patterns—correlation, distribution, group difference.
- Visualize the relationships: Use plots to review the relationship between AI use and motivation or performance measures.
Think about how those findings would be applicable in actual educational contexts: Can AI help incentivize low-energy students?
Where Python Fits into This Journey
Coming into the Digital Research Incubator, Python appeared intimidating.It seemed like the kind of language that belonged to people from a different planet. But the process of learning it has been surprisingly intuitive. At times, it’s like having a conversation with a patient computer: If I word something right, the answer comes back in plain language; if not, it allows me to try again.
The harder part is not really Python—it’s the mathematics and data. Sifting through raw, imperfect data and making something out of it is part art and part negotiation. Python’s versatility, though powerful, can also be intimidating, and I regularly contemplate what constitutes a “complete” program versus a bunch of analytical steps.
But slogging through that ambiguity at this point is part of the project—a reminder that research is seldom linear and, that motivation (mine too)is fueled by small victories.
Closing Thoughts
This blog post is all about the origins of questions, motivations, and nascent ideas. My friend’s story was the inspiration for the narrative. Motivation theory provided the language. AI research provided the wider view. And Python provided me the tools to interact with everything in a different manner. I don’t think that this project is going to resolve the tensions between AI, motivation, and learning. But I hope it can make small patterns visible, prompt better questions and shed light on what we might do to better help students who feel defeated about learning.