Imagine a classroom where every student’s unique learning journey is understood — where their strengths, areas for growth, and the process of skill-building are fully visible. Learning analytics makes this vision a reality, offering a deeper view beyond traditional grades and test scores. With these insights, teaching transforms from simply delivering content into a personalized experience that meets each student exactly where they are.
This vision drives the Turku Research Institute for Learning Analytics work, where the award-winning platform ViLLE is making personalized education a reality. Developed over the last twenty years, ViLLE has achieved remarkable success in Finland, and internationally, a similar platform is known as Eduten. Its varied curricula include a wide range of carefully designed interactive tasks that can be combined with other learning methods. Real-time analytics reveal patterns that might otherwise go unnoticed: the quiet student struggling with foundational concepts, the high-achieving student showing signs of disengagement, or a class of students excelling in one area but facing challenges in another. ViLLE also offers teachers recommendations for complementary resources, enhancing support for each student. In 2025, the development of next-generation ViLLE will begin, introducing advanced adaptive learning features led by a team of experienced academics and doctoral researchers.
In the field of informatics, learning analytics is where data science, AI, and education converge. It allows us to explore patterns in student engagement, performance, and behaviour, enabling intelligent systems that support personalized learning at scale. ViLLE has made strides in adaptive curriculum development for informatics through multi-parameterized tasks, offering students a customized learning experience. This work began with support from the Nordic Council of Ministers and is now expanding with the EU Erasmus project CT&MathAble, which explores the connections between algebraic and computational thinking.
Figure 1 – Examples of tasks in ViLLE.
The four stages of learning analytics — descriptive, diagnostic, predictive and prescriptive — build progressively to offer a comprehensive understanding of the learning process.
Descriptive analytics serves as the foundation, providing an overview of what is happening in the learning environment. By analyzing data like task points, time usage, and activity completion, educators can capture a snapshot of student progress and engagement. Moving beyond observation, diagnostic analytics delves into why these patterns occur, identifying underlying factors such as common misconceptions or specific areas where students struggle. This stage allows educators to uncover the root causes behind student performance and learning behaviors.
Figure 2 – Descriptive data from the tasks in one lesson. Answer accuracy, time usage and number of active students are visualized.
Predictive analytics then uses historical data to anticipate future outcomes, answering questions like, “What might happen next?”. For instance, predictive models can help determine which students may excel or encounter challenges, allowing educators to proactively intervene. Finally, prescriptive analytics takes learning insights to the next level by recommending targeted actions based on predicted outcomes. It guides teachers in implementing personalized strategies, such as assigning specific resources or adjusting lesson plans, to optimize learning for each student.
Figure 3 – Predictive analytics in ViLLE. Students marked with the circle are at risk of failing the course and need support from the teacher.
This background gives Turku Research Institute for Learning Analytics a strong interest in the DIGITAL FIRST project. Fresh approaches are needed to engage the “digital native” generation — young people who have grown up immersed in digital technology. The shift in language education from a structuralist to a functionalist approach offers an inspiring model, one that can help focus efforts on equipping students to become active, capable participants in the digital world. Students need to learn informatics not just by memorizing structures, but by understanding how to apply their knowledge in real-world contexts — just as language learning has shifted from a structuralist to a functionalist approach. In informatics, this shift means focusing not just on syntax and theory but on problem-solving and application. Learning analytics plays a crucial role here, supporting both the research and implementation phases by offering insights that guide and refine this transformative approach.
Figure 4 – Students practicing with ViLLE tasks.
In short, learning analytics offers teachers a way to teach smarter, not harder. It’s a tool for creating a richer, more responsive learning environment — one where every student feels seen, understood, and empowered to achieve their potential. We need solutions which ensure that students don’t just learn about informatics but learn through it, developing the critical, functional skills they’ll need in real-world applications.
